PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
SCHEDULE F, PART 2
OPERATIONAL SERVICE LEVEL
Schedule F, Service Level Requirements
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
CONTENTS
1
INTRODUCTION....................................................................................................................1
2
SERVICE LEVEL CRITERIA...............................................................................................3
3
SHADOW METRICS ............................................................................................................83
4
SERVICE SCORE .................................................................................................................84
5
IDENT1 WORKLOAD..........................................................................................................91
Schedule F, Service Level Requirements
i
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
1
INTRODUCTION
This Part of Schedule F (Part 2) details the SLR for the IDENT1 Service, except for Livescan
Services which are detailed separately in Part 3 of this Schedule.
The Schedule details the following aspects of the SLR:
(a)
Service Level Metrics and the Target Service Levels applicable from the milestones
TOR NAFIS and TOR SAFR;
(b)
the changes in Service Level Metrics to be incorporated at subsequent program
milestones as part of the continuous improvement process;
(c)
the planned use of shadow metrics to evolve the SLR.
The framework detailing how the different Service Categories interact is detailed in. Figure 1.0-1
below.
Business
Compute
TBD Metrics
Category
Category Weights
Score
Availability
Capacity
System
Map Score to
Response Time
WiSi
WcSc
Payment
Accuracy
Service Mgmt
Interfaces
User
Overall Score
Compute
TBD Metrics
Category
Score
Category Scores
Figure 1.0-1. IDENT1 Service Level Structure
This Part of the SLR will evolve in accordance with the process and procedures documented in Part 1
of this Schedule to consist of three Service Categories that address respectively IDENT1 business
values, system performance and user needs. Each Service category will contain a series of Service
Criteria with respective Service Level Metrics.
Each Service Category will generate a category score based on monthly measurements of its defined
Service Level Metrics. The Service Category scores are then weighted and summed to produce an
Overall Score that can be used in conjunction with an agreed payment profile as detailed in Schedule
E to determine the actual payment.
As Service Level Metrics within Service Categories are implemented, The Service Category weights
that will affect the scoring and payment need to be negotiated and agreed with the Authority in
accordance with Part 1 of this Schedule. The range of weighting for each Service Category is detailed
in Figure 1.0-2 below.
Schedule F, Service Level Requirements
1
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Service Category
Minimum Weight
Maximum Weight
System Performance
50%
100%
Business Values
0%
50%
User Needs
0%
20%
Figure 1.0-2. Bounds for Service Category Weights
At the milestone TOR NAFIS this SLR will become effective and the SLR will be based only on the
Service Category - System Performance. The Service Level Criteria implemented under this Service
Category at this milestone are specifically availability, capacity/throughput, response time, service
management, and search accuracy.
Schedule F, Service Level Requirements
2
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
2
SERVICE LEVEL CRITERIA
This section defines the Service Level Criteria. The Service Level Criteria are indicative measures of
the services supplied by the IDENT1 Contractor. This section defines the criteria to be measured, the
formulae used to calculate the criteria, and the scores to be associated with the actual values for each
service level criteria.
On a monthly basis, each IDENT1 site (i.e., Bureaux and IDENT1 Training Centres) will receive a
Service Score. This Service Score will be used to assess the quality of the IDENT1 Service and
computed as described in Figure 2.0-1 below and in Section 4 of this Part of this SLR. Figure 2.0-1
illustrates the steps that will be used to compute a score for each measure and how these scores will be
used to compute the Site Service Score. Specifically, the steps are as follows:
Compute Individual
Compute the MEASURE
Site Scores
Using the defined
for Each Measure
Formula
Select Site SCORE
Compute the
Corresponding
AVERAGE MEASURE
to MEASURE
Weight the SCORE
Compute Individual
Compute Site
According to Service
SERVICE SCORE
Site Service Scores
Weighting Factors
Figure 2.0-1 IDENT1 Measure Flow Diagram
Step 1 Compute individual Site SCORE for each MEASURE.
(a)
Compute each MEASURE for each Service Level Criteria using the formulae
presented in Section 2.1 through 2.5 of this SLR.
(b)
Where appropriate, compute the AVERAGE MEASURE for that month. Note that
this step does not apply to all service measures. In some instances, the MEASURE
provides the singular value used to select the SCORE.
(c)
Select the SCORE that corresponds to the computed MEASURE.
Step 2 Compute individual Site SERVICE SCORE.
(a)
WEIGHT each SCORE by using the weighting factor assigned to each MEASURE.
This weight is assigned as part of this SLR.
(b)
Compute the SERVICE SCORE by taking the summation of the product of each
SCORE and its WEIGHT. This is shown in Equation 2.0-1.
Schedule F, Service Level Requirements
3
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Equation 2.0-1:
Service Score = ∑ (Score × Weight )
The computation of an individual Site SCORE for each MEASURE (step 1 above) is presented in
Section 2.1 through 2.5 of this SLR.
2.1
Operational Availability
2.1.1
Scope
This service level criterion defines metrics for the measurement of operational
availability: availability of the IDENT1 service reflecting loss of service due to
unplanned downtime. The service is defined in terms of serviceability of critical
equipment.
2.1.2
Operational Availability Calculation
Operational Availability (Ao) will include the critical equipment, as defined in
Section 2.1.3.1, 2.1.4.1, and 2.1.5.1. These characteristics will result in an Ao
calculation for the Central Segment, Bureaux Segment and Training System(s) based
upon total hours available monthly and downtime of unplanned interruptions. Ao will
be calculated to two decimal places using Equation 1:
( m
H − DTm)
Equation 1:
Ao =
×100
m
H
Where,
(a)
Hm will be defined as the total hours in a given month. This number will be
determined each month as 24 hours multiplied by the number of days in the
month.
(b)
DTm will be defined as the unplanned downtime, i.e., the total number of
hours in a given month that critical equipment is not available.
The IDENT1 Service will be available 24 hours per day, every day of the year. The
exceptions to this will be
(a)
Planned system interruptions of up to 4 hours in any one day, to take place
contiguously between the hours of 22:00 and 06:00. These planned
interruptions may affect only a part of the service and will be planned at least
four (4) weeks ahead of the scheduled time of the disruption with prior
agreement between Authority, the relevant Bureaux and Contractor, or as
agreed between the parties. The interruptions are described in terms of which
of the described services will be impaired and to what degree.
(b)
Planned local outages of no more than 2 hours per day on any one
workstation or local server to enable the switch to new versions of software
to be achieved in an orderly fashion. These short outages will take place
between the hours of 22:00 and 06:00 unless other arrangements are made
between the Bureaux and the Contractor. The Contractor will notify the users
at least two (2) weeks before these types of outages occur, or as agreed
between the Contractor and the Site Point of Contact.
Schedule F, Service Level Requirements
4
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
(c)
Emergency outages at any time of the day, by mutual agreement with the
Contractor and the Authority or Site Point of Contact, to solve severe
operational problems. Examples of such emergency outages may include, but
may not be limited to: fire in the facilities, unexpected power outage within
the central computer facility, or severe system malfunction that could
jeopardise the National collections.
(d)
Other exceptions as defined in Part 1 of this SLR.
In the event that an emergency operational need arises before or during an
interruption, the interruption may be postponed or service restored by the Contractor
as soon as practical so as to allow IDENT1 services to operate.
For multiple, redundant items at site, downtime will commence when all of a given
item are no longer available. For example, multiple, redundant processors are used
within select Bureaux, ITMCS, AFRS, and ISRS. Downtime will not commence
until all servers within the subsystem are unavailable, thereby halting normal
operations. The same is true for fingerprint matcher banks. Downtime will
commence when all banks of a given search type (e.g., Print-to-Print, etc.) are
unavailable. Degraded performance as a result of a failure of a single item will not be
considered downtime. However, such degraded performance may impact bureau
operations and will be reflected in the throughput and system performance service
level criteria as defined in Section 2.2 and 2.3 of this SLR.
Downtime will start and stop under the authorisation of the IDENT1 Help Desk and
in consultation with the Site Point of Contact. Downtime will only include unplanned
interruptions. Downtime is composed of administrative delays associated with co-
ordinating personnel arrival time at a site (i.e., bureau personnel available for site
technician to enter a site), travel time to the site (if applicable), service restoration
time, and time to test corrective actions. Downtime excludes any bureau-originated
administrative delays that impede service delivery or restoration of service.
Planned interruptions will be defined as all interruptions of services that take place as
scheduled, and with prior agreement between the affected Head of Bureau(x) and the
Contractor. Planned interruptions shall exclude any period of unavailability required
to fix problems of functionality or throughput, which result from the Contractor’s
failure to meet IDENT1 requirements, irrespective of whether or not these
interruptions are scheduled and undertaken with the prior agreement of the Authority
or the bureau(x) involved
Planned Downtime will not reduce the hours available per month (Hm) nor increase
the unplanned downtime per month (DTm). In all instances, the Contractor will
declare the duration of a Planned Downtime Period, as defined above, prior to starting
Planned Downtime. Planned Downtime in excess of the Planned Downtime Period
will become Unplanned Downtime commencing at the end of the period.
When a critical component or all of the system becomes unavailable, the Contractor
will inform the Bureaux and will continue to manage the Service in such a way as to
ensure that the daily workload is completed within the working day. If this becomes
impractical, the problem will be escalated within the Contractor for decisions on
priorities, in accordance with the IDENT1 Maintenance Plan.
Provisional Handback - Upon completion of significant work on the system, such as
an ECP, Northrop Grumman may request a period of time during which the normal
conditions of the SLR continue to be suspended. This time period will be referred to
Schedule F, Service Level Requirements
5
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
as Provisional Handback of the system. The purpose of this is to ensure that there is
no adverse impact of the change once the full operational workload is submitted to
the system. Application of Provisional Handback will be agreed in advance with
PITO, including a maximum duration.
The Disaster Recovery System will be available to support IDENT1 operations in the
event of a disaster at the primary site. The DRS must be fully operational within 24
hours of a requested start. Those 24 hours will be considered planned down time.
Time taken beyond 24 hours will be considered unplanned down time.
2.1.3
Central Segment Availability
2.1.3.1 Central Segment Critical Components
The Central Segment consists of equipment at the Primary Site at Hendon and at the
DRS site. Central Segment Ao will be based on the total hours in a given month and
unplanned downtime of the critical equipment in a given month at each of the sites.
Critical and Non-Critical items for the Primary Site are identified in Figure 2.1-1 and
for the DRS in Figure 2.1-2.
Critical Components Primary Site
Non-Critical Components
Servers (ITMCS-ISRS-AFRS)
Server Tape Jukebox (ITMCS-ISRS-AFRS)
Storage Controllers (ISRS)
Small SCSI Switch (ITMCS-ISRS-AFRS)
Storage Controller Tape Jukeboxes (ISRS)
Network Routers
Server monitor, keyboard, and mouse (ITMCS-
ISRS-AFRS)
Text Printers
Network Switches
Computer Operator Workstations
Print-to-Print Search Engines
Central Segment Personal Computer
Print-to-Mark Search Engines
Network Management Server
Mark-to-Print Search Engines
Management Console Workstation
Mark-to-Mark Search Engines
Management LAN
Central Segment RAID Assemblies (ITMCS-
HSM Server
ISRS-AFRS)
NIS+/LDAP/DNS servers
Figure 2.1-1 Primary Site Critical and Non-Critical Components
Schedule F, Service Level Requirements
6
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Critical Components – Disaster Recovery Site Non-Critical Components
Servers (ITMCS-ISRS-AFRS)
Server Tape Jukebox (ITMCS-ISRS-AFRS)
Storage Controllers (ISRS)
Small SCSI Switch (ITMCS-ISRS-AFRS)
Storage Controller Tape Jukeboxes (ISRS)
Network Routers
Server monitor, keyboard, and mouse (ITMCS,
ISRS, AFRS)
Text Printers
Network Switches
Computer Operator Workstations
Print-to-Print search engines
DRS Personal Computer
Print-to-Mark search engines
Network Management Server
Mark-to-Print search engines
Management Console Workstation
Mark-to-Mark search engines
Management LAN
Replicated RAID Assemblies (ITMCS-ISRS-
HSM Server
AFRS)
NIS+/LDAP/DNS servers
Figure 2.1-2. Central Segment - DRS Critical and Non-Critical Components
2.1.3.2 Central Segment Ao Scoring
The Central Segment Primary Site and DRS Site Ao value will be computed monthly
using Equation 1 of Section 2.1.2.
Once each Ao value has been computed, the SCORE will be selected from the
Scoring Table presented as Figure 2.1-4 to produce the Central Ao Service Level
Scores as shown in Figure 2.1-3.
Ao
SCORE
Central Ao Primary
Central Ao DRS
Figure 2.1-3 Central Segment Ao Score
Schedule F, Service Level Requirements
7
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Central
SCORE Central
SCORE Central
SCORE
Segment
Segment
Segment
Ao
Ao
Ao
100.0
110.0
98.3
93.0
96.6
76.0
99.9
109.0
98.2
92.0
96.5
75.0
99.8
108.0
98.1
91.0
96.4
74.0
99.7
107.0
98.0
90.0
96.3
73.0
99.6
106.0
97.9
89.0
96.2
72.0
99.5
105.0
97.8
88.0
96.1
71.0
99.4
104.0
97.7
87.0
96.0
70.0
99.3
103.0
97.6
86.0
95.9
69.0
99.2
102.0
97.5
85.0
95.8
68.0
99.1
101.0
97.4
84.0
95.7
67.0
99.0
100.0
97.3
83.0
95.6
66.0
98.9
99.0
97.2
82.0
95.5
65.0
98.8
98.0
97.1
81.0
95.4
64.0
98.7
97.0
97.0
80.0
95.3
63.0
98.6
96.0
96.9
79.0
95.2
62.0
98.5
95.0
96.8
78.0
95.1
61.0
98.4
94.0
96.7
77.0
≤95.0
60.0
Figure 2.1-4 Central Segment Ao Scoring Table
The Central Ao Service Level Scores will be allocated to each IDENT1 site on an
equal basis, with each Site receiving the same value, in the Central Ao columns of
Figure 4.1-1.
In the event of a disaster at the primary site and the use of the DRS to support
IDENT1 operations, no Central Ao - Primary will be computed. Rather the Ao of the
DRS site will be used as the Central Ao - Primary value in Figure 4.1-1 with a weight
of 0.10, and the value of the DRS Ao will be set to zero (0). If there is a disaster at the
DRS, the Central Ao - Primary value will be computed as normal but with a weight of
0.10 and a value of zero (0) will be used for the DRS Ao.
2.1.4
Bureau Segment Availability
2.1.4.1 Bureau Segment Critical Components
Schedule F, Service Level Requirements
8
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Bureau Segment Ao will be based on the total hours in a given month and unplanned
downtime of the critical equipment in a given month. Critical and Non-Critical items
are identified in Figure 2.1-5.
Where a site has multiple instances of the same critical equipment item, as defined in
Figure 2.1-5, (e.g., workstation(s)), the Contractor and the Head of Bureau(x) (or
designated nominee) will determine whether or not the site is unavailable to process
its work in order to compute a monthly downtime for that site. The results of this
consultation will be recorded in the Help Desk Management System. At all sites, the
text printer will be non-critical. However, where a site has a text printer and a
graphics printer, the graphics printer may be critical if the Contractor and the Head of
Bureau(x) (or designated nominee) determine that the site is unavailable to process its
work without the availability of that printer. The results of this consultation will also
be recorded in the Help Desk Management System.
The External Transaction Bureau (ETB) server is physically located at Central.
However, it is considered a Bureau just as the Hendon Bureau is considered a Bureau.
The critical component for the ETB bureau is the Transaction Server. This Bureau
does not have printers or workstations.
Critical Components
Non-Critical Components
WAN
Tape Jukebox
Server(s)
Text Printers
Workstation(s)
Server monitor, keyboard, mouse, CD ROM and
tape drive
Mark and Ten Print Scanner(s)
Graphic printer(s) unless agreed critical per left
column
Network Router
Network Hub
Network Switch
Bar Code Printer(s)
Bar Code Reader(s)
RAID Assembly
PNC
Satellcom Gateways
Graphic Printer(s) under agreed exception
conditions
Operational Response Matchers
Figure 2.1-5 Bureau Segment Critical and Non-Critical Components
Schedule F, Service Level Requirements
9
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
2.1.4.2 Bureau Segment Ao Scoring
The Ao value for each Bureau will be computed monthly using Equation 1 of Section
2.1.2.
Schedule F, Service Level Requirements
10
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Bureau
Ao
SCORE
Bureau
Ao
SCORE
Avon + Somerset
Norfolk
Bedfordshire
North Wales
British Transport
North Yorkshire
Police
Northamptonshire
Cambridgeshire
Northumbria
Cheshire
Nottinghamshire
City of London
South Wales
Cleveland
South Yorkshire
Cumbria
Staffordshire
Customs and Excise
Suffolk
Derbyshire
Surrey
Devon + Cornwall
Sussex
Dorset
Thames Valley
Durham
Warwickshire
Dyfed Powys
West Mercia
Essex
West Midlands
Gloucestershire
West Yorkshire
Greater Manchester
Wiltshire
Gwent
Ext. Transaction
Hampshire
Bureau
Hertfordshire
Figure 2.1-6 Bureau Segment Ao Score
Humberside
Kent
Lancashire
Leicestershire
Lincolnshire
Merseyside
Metropolitan
N.I.S./Residual Bureau
Schedule F, Service Level Requirements
11
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Once the Bureau Ao value has been computed for each Bureau, the Service Level
Score for each Bureau will be selected from the Scoring Table presented as Figure
2.1-7 to produce the Bureau Ao Service Level Scores as shown in Figure 2.1-6.
Bureau
SCORE Bureau
SCORE Bureau
SCORE
Segment
Segment
Segment
Ao
Ao
Ao
100.0
110.0
98.3
93.0
96.6
76.0
99.9
109.0
98.2
92.0
96.5
75.0
99.8
108.0
98.1
91.0
96.4
74.0
99.7
107.0
98.0
90.0
96.3
73.0
99.6
106.0
97.9
89.0
96.2
72.0
99.5
105.0
97.8
88.0
96.1
71.0
99.4
104.0
97.7
87.0
96.0
70.0
99.3
103.0
97.6
86.0
95.9
69.0
99.2
102.0
97.5
85.0
95.8
68.0
99.1
101.0
97.4
84.0
95.7
67.0
99.0
100.0
97.3
83.0
95.6
66.0
98.9
99.0
97.2
82.0
95.5
65.0
98.8
98.0
97.1
81.0
95.4
64.0
98.7
97.0
97.0
80.0
95.3
63.0
98.6
96.0
96.9
79.0
95.2
62.0
98.5
95.0
96.8
78.0
95.1
61.0
98.4
94.0
96.7
77.0
≤95.0
60.0
Figure 2.1-7 Bureau Ao Scoring Table
The Bureau Segment Ao Service Level Score will be placed in the Site Ao column of
Figure 4.1-1 for each IDENT1 Bureau.
2.1.5
Training System Availability
2.1.5.1 Training System Critical Components
Training System Ao will be based on the total hours in a given month and unplanned
downtime of the critical equipment in a given month. Critical and Non-Critical items
are identified in Figure 2.1-8.
Schedule F, Service Level Requirements
12
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
When a critical equipment item, as defined in Figure 2.1-8, (e.g., workstation(s)),
becomes unavailable at a Training Centre, the Contractor and the Head of the
Training Centre (or designated nominee) will assess the impact on that centre’s
course schedule. If, because of a loss of an equipment item, the Training Centre is
unable to continue with its training schedule, then the time that the item is unavailable
will be recorded as downtime in order to compute a monthly downtime for that site.
If the loss of an equipment item does not affect the Training Centre’s training
schedule, then no downtime will be recorded until such time as the loss affects the
training schedule. The results of this consultation will be recorded in the Help Desk
Management System. The text printer will be non-critical at the Training Centres.
However, where a site has a text printer and a graphics printer, the graphics printer
may be deemed critical if the Contractor and the Head of the Training Centre (or
designated nominee) determine that the Training Centre is unable to continue with its
training schedule without it. The results of this consultation will also be recorded in
the Help Desk Management System.
Critical Components
Non-Critical Components
Server(s) (combined ITMCS, ISRS, AFRS, and
Tape Jukebox
Bureau)
Workstation(s)
Text Printers
Mark and Ten Print Scanner(s)
Server monitor, keyboard, mouse, CD ROM and
tape drive
Network Router
PNN
Network Hub
Graphic Printer(s) unless agreed critical
Network Switch
Bar Code Printer(s)
Bar Code Reader(s)
Redundant Array of Independent Disks (RAID)
assembly
Graphic Printer(s) under agreed exception
conditions
Figure 2.1-8 Training System Critical and Non-Critical Components
2.1.5.2 Training System Ao Scoring
The Ao value for each Training System will be computed using Equation 1 of Section
2.1.2
Schedule F, Service Level Requirements
13
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Training System Ao
SCORE
Hendon
Durham
Figure 2.1-9. Training System Ao Score
Once the Ao value has been computed for each Training System, the SCORE for each
Training System will be selected from the Scoring Table presented as Figure 2.1-10
to produce the Training System Ao Service Level Scores as shown in Figure 2.1-9.
Training SCORE Training SCORE Training SCORE
Segment
Segment
Segment
Ao
Ao
Ao
100.0
110.0
98.3
93.0
96.6
76.0
99.9
109.0
98.2
92.0
96.5
75.0
99.8
108.0
98.1
91.0
96.4
74.0
99.7
107.0
98.0
90.0
96.3
73.0
99.6
106.0
97.9
89.0
96.2
72.0
99.5
105.0
97.8
88.0
96.1
71.0
99.4
104.0
97.7
87.0
96.0
70.0
99.3
103.0
97.6
86.0
95.9
69.0
99.2
102.0
97.5
85.0
95.8
68.0
99.1
101.0
97.4
84.0
95.7
67.0
99.0
100.0
97.3
83.0
95.6
66.0
98.9
99.0
97.2
82.0
95.5
65.0
98.8
98.0
97.1
81.0
95.4
64.0
98.7
97.0
97.0
80.0
95.3
63.0
98.6
96.0
96.9
79.0
95.2
62.0
98.5
95.0
96.8
78.0
95.1
61.0
98.4
94.0
96.7
77.0
≤95.0
60.0
Figure 2.1-10 Training System Ao Scoring Table
Schedule F, Service Level Requirements
14
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
The Training System Ao Service Level Score will be placed in the Site Ao column of
Figure 4.1-1 for each IDENT1 Training Site.
2.1.6
Future Changes for Availability
Several changes to the availability service level criterion will be incorporated by no
later than FOC. These include the following:
(a)
Updating the score tables to reflect score of 100 for measured
availability Target Service Level of 99.5% at FOC and 99.6% at
FOC+3 years
(b)
Updating the Critical and Non-Critical component list to reflect
IDENT1 Service architecture and hardware
(c)
Deletion of separate scoring for Primary Central and DRS sites
(d)
Implementation of Planned Downtime metrics
2.1.6.1 Scoring Table Update
To reflect the increase of availability requirements from 99% for NAFIS applicable at
TOR to 99.5% for the IDENT1 Service applicable at FOC when the full architecture
is deployed, availability score tables for both the Central site and the Bureaux will be
updated to provide a score of 100 when the Target Service Level for availability of
99.5% is met. The updated table, Figure 2.1-11, applicable for the Central site, the
Bureaux and the training sites is shown below:
Ao
Score
Ao
Score
Ao
Score
Ao
Score
Ao
Score
100.0
110.0
99.20
97.0
98.20
84.0
97.55
71.0
96.90
58.0
99.95
109.0
99.10
96.0
98.15
83.0
97.50
70.0
96.85
57.0
99.90
108.0
99.00
95.0
98.10
82.0
97.45
69.0
96.80
56.0
99.85
107.0
98.90
94.0
98.05
81.0
97.40
68.0
96.75
55.0
99.80
106.0
98.80
93.0
98.00
80.0
97.35
67.0
96.70
54.0
99.75
105.0
98.70
92.0
97.95
79.0
97.30
66.0
96.65
53.0
99.70
104.0
98.60
91.0
97.90
78.0
97.25
65.0
96.60
52.0
99.65
103.0
98.50
90.0
97.85
77.0
97.20
64.0
96.55
51.0
99.60
102.0
98.45
89.0
97.80
76.0
97.15
63.0
96.50
50.0
99.55
101.0
98.40
88.0
97.75
75.0
97.10
62.0
99.50
100.0
98.35
87.0
97.70
74.0
97.05
61.0
99.40
99.0
98.30
86.0
97.65
73.0
97.00
60.0
99.30
98.0
98.25
85.0
97.60
72.0
96.95
59.0
Figure 2.1-11. Central Site and Bureaux Ao Scoring Table at FOC
Schedule F, Service Level Requirements
15
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
To provide continuous improvement, the target value for availability will be further
increased to 99.6% at FOC+3 years. At that milestone, the scores shown in Figure
2.1-11 will be shifted upwards to reflect a score of 100 for a measured Ao of 99.6%
for the Central, Bureaux, and training sites. Further improvement in availability will
be incentivised through the SLR scoring structure.
2.1.6.2 Critical Equipment
To support the availability measurement, the tables defining the critical and non-
critical component lists for the Central site and the Bureaux will be updated at FOC.
The updated tables will fully reflect the deployed IDENT1 Service architecture and
hardware.
2.1.6.3 Deletion of DRS
At TOR, availability is measured on the NAFIS architecture with a Primary Central
Site and a DRS. Separate measurements and scoring are performed for these sites
with a weight of 7% for the Primary and 3% for the DRS. With the full deployment of
the IDENT1 web service based architecture with load balancing at FOC, separation
scoring becomes unnecessary due to the symmetry of the architecture. A single
availability measurement of the central sites will be made with a weight of 10%.
2.1.6.4 Measurement of Planned Downtime
To support the goal of minimizing planned downtime in addition to unplanned
downtime, shadow metrics will be defined and evaluated. The goal is for a single
metric that incorporates both planned and unplanned downtime. As a minimum,
separate metrics on planned downtime will be defined, agreed with the Authority and
evaluated as shadow metrics prior to the FOC. These metrics will have target values
derived from the DOR requirements of total planned downtime of 4 hours in a rolling
24-hour window, and frequency of occurrence of no more than two in a rolling 4-
week period for the central services; and total planned downtime of 4 hours in a
rolling 4-week window for each Bureau. If supported by the evaluation results, it is
envisioned that these metrics will be implemented per the SLR change process
defined in Part 1. The recommended weighting structure incorporating the planned
downtime metrics is shown in Figure 2.1-12.
Availability Metrics and Weights
Central
Bureaux
Ao
Planned Downtime
Ao
Planned Downtime
0.075
0.025
0.075
0.025
Figure 2.1-12. Availability Metrics and Weights
2.1.6.5 Scottish Bureaux
In addition to the above changes, at intermediate milestones prior to the FOC, i.e.,
when the Scottish Bureaux become operational, availability will be measured and
scored for these Bureaux in a similar manner to existing Bureaux using the then
applicable metrics and scoring tables.
Schedule F, Service Level Requirements
16
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
2.2
Throughput
2.2.1
Scope
Ten Print and Mark processing will be measured in terms of throughput. Throughput will be
defined as the percentage of Ten Print or Print Set and Mark transactions completed within a
specified time period. These criteria will measure the throughput of the system while not
restricting working practices. This section is subject to design limitations expressed in
Section 5 of this SLR, Workload Assumptions.
2.2.2
Ten Print Throughput
Ten Print Throughput will be computed for each Bureau and the NIS. Ten Print Throughput
will not be computed for the Training Centres.
All processing steps within Ten Print processing, from Ten Print form take-on through Ten
Print close-out, will be included in the Ten Print Throughput measure as a means of capturing
the overall throughput of the system. For this reason, Ten Print Throughput will be measured
on an aggregate monthly basis and will not consider other factors affecting throughput such as
working practices, search priority and search scope.
Equation 2.2-1 will be used to calculate the Bureau Ten Print Throughout (TPt).
TPc
Equation 2.2-1:
TPt = MIN(TPa TP ) ×100
,
max
where
(a)
TPc is the total number of Ten Print transactions processed through close-out
during the month. Close-out is the final disposition of all Ten Print
transactions.
(b)
TPa is the total number of Ten Print forms received and processed during the
month. This will be measured at the point of submission.
(c)
TPmax is the monthly number of Ten Prints received and processed as defined
in Section 5 of this SLR for the respective bureau.
In Equation 2.2-1, the denominator will be the lesser of either TPa or TPmax. Ten Print
Throughput for each Bureau will be computed using Equation 2.2-1. This value will be the
TEN PRINT THROUGHPUT value as shown in column 2 of Figure 2.2-1)
Transactions for the External Transactions Bureau will be considered closed out when the
response to a search request has been transmitted to the requesting external system.
Schedule F, Service Level Requirements
17
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Bureau
Ten Print
SCORE
Bureau
Ten Print
SCORE
Throughput
Throughput
Avon + Somerset
Norfolk
Bedfordshire
North Wales
British Transport Police
North Yorkshire
Cambridgeshire
Northamptonshire
Cheshire
Northumbria
City of London
Nottinghamshire
Cleveland
N.I.S./Residual Bureau
Cumbria
South Wales
Customs and Excise
South Yorkshire
Derbyshire
Staffordshire
Devon + Cornwall
Suffolk
Dorset
Surrey
Durham
Sussex
Dyfed Powys
Thames Valley
Essex
Warwickshire
Gloucestershire
West Mercia
Greater Manchester
West Midlands
Gwent
West Yorkshire
Hampshire
Wiltshire
Hertfordshire
Ext. Transaction Bureau
Humberside
Figure 2.2-1 Ten Print Throughput Scores
Kent
Lancashire
Leicestershire
Lincolnshire
Merseyside
Metropolitan
N.I.S./Residual Bureau
Schedule F, Service Level Requirements
18
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
In the event that no applicable ten prints are processed by the site in the reporting period, a fixed
SCORE of 100 shall be applied to TEN PRINT THROUGHPUT for that site. Once the TEN PRINT
THROUGHPUT value has been computed for each Bureau, the SCORE for each Bureau will be
selected from the Scoring Table presented as Figure 2.2-2.
Ten Print
Ten Print
Ten Print
Ten Print
Ten Print
Score
Score
Score
Score
Score
Throughput
Throughput
Throughput
Throughput
Throughput
≥110
115
99
99
88
88
77
77
66
66
109
113
98
98
87
87
76
76
65
65
108
111
97
97
86
86
75
75
64
64
107
109
96
96
85
85
74
74
63
63
106
107
95
95
84
84
73
73
62
62
105
105
94
94
83
83
72
72
61
61
104
104
93
93
82
82
71
71
≤60
60
103
103
92
92
81
81
70
70
102
102
91
91
80
80
69
69
101
101
90
90
79
79
68
68
100
100
89
89
78
78
67
67
Figure 2.2-2 Ten Print Throughput Scoring Table
The SCORE will be used for SLR reporting as shown in Figure 4.1-1. Ten Print Throughput
will be reported on an individual Bureau basis.
2.2.3
Mark Throughput
Mark Throughput will be computed for each Bureau. Mark Throughput will not be
computed for the Training Centres.
Mark Throughput will be measured on an aggregate monthly basis and will not
consider other factors affecting throughput such as working practices, search priority
and search scope.
Equation 2.2-2 will be used to calculate the Bureau Mark Throughput (Mt).
( ni
M +
i
M + Mr + Mpni + Mpi)
Equation 2.2-2:
t
M = MIN(MPa MP )
×100
,
max +
pr
M
where
(a)
Mni is defined as the number of Mark-to-Print non-identification decisions
made at the Bureau during the month as the result of a Mark-to-Print search
submission. This decision is made after all images have been down-loaded to
Schedule F, Service Level Requirements
19
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
the Bureau and the Fingerprint Officer has reviewed the respondent list to
make the first-level identification determination. Second- and third-level
identifications are excluded from the calculation as the Bureau may take
some time to process the Mark through third-level identification.
(b)
Mi is defined as the number of Mark-to-Print identification decisions made at
the Bureau during the month as the result of a Mark-to-Print search
submission. This decision is made after all images have been down-loaded to
the Bureau and the Fingerprint Officer has reviewed the respondent list to
make the first-level identification determination. The remainder of the
respondents after the Mark-to-Print identification is made are excluded from
the calculation.
(c)
Mr is defined as the remainder of searches that remain in ‘In Compare 1’ four
months after those searches were submitted. For example, searches that were
launched and returned in January, which had not been idented or non-idented
by May, would be Mr in the Mark Throughput calculation for May. If these
searches were subsequently idented or non-idented, the statistical collection
process will prevent them from being included in the calculation a second
time.
(d)
MPa is defined as the actual number of Mark-to-Print searches submitted at
the Bureau during the month. Note that a re-launch or repeat of a search is
considered a separate search.
(e)
MPmax is defined as the maximum number of Mark-to-Print searches
submitted at the Bureau during the month. The maximum number of Mark-
to-Print searches is defined, for each Bureau, in Section 5 of this SLR.
(f)
Mpni is defined as the number of Mark-to-Police Elimination Database non-
identification decisions made at the Bureau during the month as the result of
a Mark-to-Police Elimination Database search submission. This decision is
made after all images have been down-loaded to the Bureau and the
Fingerprint Officer has reviewed the respondent list to make the first-level
identification determination.
(g)
Mpi is defined as the number of Mark-to-Police Elimination Database
identification decisions made at the Bureau during the month as the result of
a Mark-to-Police Elimination Database search submission. This decision is
made after all images have been down-loaded to the Bureau and the
Fingerprint Officer has reviewed the respondent list to make the first-level
identification determination.
(h)
Mpr is defined as the number of Mark-to-Police Elimination Database
searches that are returned with one or more respondents.
In Equation 2.2-2, the denominator will be the lesser of either MPa or MPmax plus Mpr. Mark
Throughput for each Bureau will be computed using Equation 2.2-2. This value will be the
MARK THROUGHPUT value as shown in column 2 of Figure 2.2-3.
Schedule F, Service Level Requirements
20
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Bureau
Mark
SCORE
Bureau
Mark
SCORE
Throughput
Throughput
Avon + Somerset
N.I.S./Residual Bureau
Bedfordshire
Norfolk
British Transport Police
North Wales
Cambridgeshire
North Yorkshire
Cheshire
Northamptonshire
City of London
Northumbria
Cleveland
Nottinghamshire
Cumbria
South Wales
Customs and Excise
South Yorkshire
Derbyshire
Staffordshire
Devon + Cornwall
Suffolk
Dorset
Surrey
Durham
Sussex
Dyfed Powys
Thames Valley
Essex
Warwickshire
Gloucestershire
West Mercia
Greater Manchester
West Midlands
Gwent
West Yorkshire
Hampshire
Wiltshire
Hertfordshire
Ext. Transaction Bureau
Humberside
Figure 2.2-3 Mark Throughput Score
Kent
Lancashire
Leicestershire
Lincolnshire
Merseyside
Metropolitan
.
-
-
Schedule F, Service Level Requirements
21
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
In the event that no applicable marks are processed by the site in the reporting period, a fixed
SCORE of 100 shall be applied to MARK THROUGHPUT for that site. Once the MARK
THROUGHPUT value has been computed for each Bureau, the SCORE for each Bureau will
be selected from the Scoring Table presented as Figure 2.2-4. The Immigration and
Nationality Directorate is not expected to submit any marks and the External Transaction
Bureau score will be set to 100.
Mark
Mark
Mark
Mark
SCOR
Mark
SCORE
SCORE
SCORE
SCORE
Throughput
Throughput
Throughput
Throughput
E
Throughput
≥110
115
99
99
88
88
77
77
66
66
109
113
98
98
87
87
76
76
65
65
108
111
97
97
86
86
75
75
64
64
107
109
96
96
85
85
74
74
63
63
106
107
95
95
84
84
73
73
62
62
105
105
94
94
83
83
72
72
61
61
104
104
93
93
82
82
71
71
≤60
60
103
103
92
92
81
81
70
70
102
102
91
91
80
80
69
69
101
101
90
90
79
79
68
68
100
100
89
89
78
78
67
67
Figure 2.2-4 Mark Throughput Scoring Table
The SCORE will be used for SLR reporting as shown in Figure 4.1-1. Mark Throughput will
be reported on an individual Bureau basis.
2.2.4
Future Changes for Throughput/Capacity
The primary changes for the capacity/throughput service level criterion after TOR are
threefold:
(a)
The yearly changes of the target values for monthly maximum Print
Set submission (TPmax) and the mark-to-print searches (MPmax) for
each Bureau
(b)
Measurement and scoring of the Scottish Bureaux when they go live
(c)
Inclusion of Palm marks in mark-to-print throughput measurement
when palm search capability becomes available
In addition, the throughput scoring tables will be extended to reflect a minimum
allowable score of 50 by FOC.
2.2.4.1 Throughput TPmax and MPmax Parameters
Schedule F, Service Level Requirements
22
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
As previously described, TPmax and MPmax are key parameters for calculating the
throughput metrics for Print Set submissions and mark-to-print searches respectively.
TPmax and MPmax for each Bureau will be calculated for each year using the DOR
tables specified in Annex A. In particular, Tables 2.13 and 2.14 will be used for Print
Sets submissions. Figure 2.2-5 below provides the TPmax table calculated for each
Bureau for the year 2004 through 2006.
Schedule F, Service Level Requirements
23
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Monthly Ten Print Submission (MPmax)
Bureaux
Distribution
2004
2005
2006
1. Avon & Somerset
2.30%
5,520
5,796
6,072
2. Bedfordshire
2.60%
6,240
6,552
6,864
3. Cambridgeshire
0.90%
2,160
2,268
2,376
4. Cheshire
1.40%
3,360
3,528
3,696
5. City of London
0.20%
480
504
528
6. Cleveland
1.20%
2,880
3,024
3,168
7. Cumbria
1.00%
2,400
2,520
2,640
8. Derbyshire
1.40%
3,360
3,528
3,696
9. Devon &
2.30%
5,520
5,796
6,072
Cormwall
10. Dorset
0.80%
1,920
2,016
2,112
11. Durham
1.20%
2,880
3,024
3,168
12. Dyfed Powys
1.30%
3,120
3,276
3,432
13. Essex
1.90%
4,560
4,788
5,016
14. Gloucestershire
0.80%
1,920
2,016
2,112
15. Greater
5.20%
12,480
13,104
13,728
Manchester
16. Gwent
1.20%
2,880
3,024
3,168
17. Hampshire
3.50%
8,400
8,820
9,240
18. Hertfordshire
1.20%
2,880
3,024
3,168
19. Humberside
1.70%
4,080
4,284
4,488
20. Kent
2.90%
6,960
7,308
7,656
21. Lancashire
2.40%
5,760
6,048
6,336
22.Leicestershire
1.10%
2,640
2,772
2,904
23. Lincolnshire
0.90%
2,160
2,268
2,376
24. Merseyside
2.80%
6,720
7,056
7,392
25. Metropolitan
9.30%
22,320
23,436
24,552
26. Norfolk
1.10%
2,640
2,772
2,904
27. North Wales
1.00%
2,400
2,520
2,640
Schedule F, Service Level Requirements
24
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
28. North Yorkshire
1.00%
2,400
2,520
2,640
29.
1.10%
2,640
2,772
2,904
Northhamptonshire
30. Northumbria
4.10%
9,840
10,332
10,824
31. Nottinghamshire
2.10%
5,040
5,292
5,544
32. South Wales
2.60%
6,240
6,552
6,864
33. South Yorkshire
3.00%
7,200
7,560
7,920
34. Staffordshire
1.50%
3,600
3,780
3,960
35. Suffolk
0.80%
1,920
2,016
2,112
36. Surrey
1.00%
2,400
2,520
2,640
37. Sussex
1.90%
4,560
4,788
5,016
38. Thames Valley
3.00%
7,200
7,560
7,920
39. Warwickshire
0.80%
1,920
2,016
2,112
40. West Mercia
1.60%
3,840
4,032
4,224
41. West Midlands
6.60%
15,840
16,632
17,424
42. West Yorkshire
4.40%
10,560
11,088
11,616
43. Wiltshire
0.70%
1,680
1,764
1,848
44. Aberdeen
0
0
0
45. Dundee
Included in Glasgow below
0
0
0
46. Edinburgh
0
0
0
47. Glasgow
6.00%
14,400
15,120
15,840
48. British
0.60%
1,440
1,512
1,584
Transport Police
49. HM Customs &
0.10%
240
252
264
Excise
50. National
3.50%
8,400
8,820
9,240
Identification
Service
51. National Crime
0
0
0
Squad
52. Specialist
0
0
0
Crimes Bureau
Total (Upper
240,000
252,000
264,000
Bound)
Figure 2.2-5 TPmax per Bureau
Schedule F, Service Level Requirements
25
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
To account for minor variations from the DOR specified distribution, starting in year 2007,
TPmax will be based on the actual distribution of the Print Set submissions between bureaux
measured during a 12 month window in the year prior to the preceding year. For each bureau,
the average monthly Print Set submissions will be calculated. These averages will then be
used to compute the actual distribution between bureaux subject to the constraint that the total
Print Set submission load is as specified in DOR Table 2.13.
MPmax for mark-to-print searches are calculated using DOR Tables 2.17, 2.18 and 2.19. The yearly
values for each bureau are shown in Figure 2.2-6.
Monthly Mark-to-Print Searches (MPmax)
Bureaux
Distribution
2004
2005
2006
1. Avon & Somerset
2.60%
4,872
5,278
5,685
2. Bedfordshire
2.90%
5,434
5,887
6,341
3. Cambridgeshire
1.20%
2,248
2,436
2,624
4. Cheshire
0.80%
1,499
1,624
1,749
5. City of London
0.10%
187
203
219
6. Cleveland
0.70%
1,312
1,421
1,531
7. Cumbria
1.30%
2,436
2,639
2,842
8. Derbyshire
1.00%
1,874
2,030
2,186
9. Devon & Cormwall
1.70%
3,185
3,451
3,717
10. Dorset
1.50%
2,811
3,045
3,280
11. Durham
0.70%
1,312
1,421
1,531
12. Dyfed Powys
1.10%
2,061
2,233
2,405
13. Essex
1.40%
2,623
2,842
3,061
14. Gloucestershire
1.20%
2,248
2,436
2,624
15. Greater Manchester
6.00%
11,242
12,181
13,119
16. Gwent
1.10%
2,061
2,233
2,405
17. Hampshire
2.50%
4,684
5,075
5,466
18. Hertfordshire
1.10%
2,061
2,233
2,405
19. Humberside
1.30%
2,436
2,639
2,842
20. Kent
2.70%
5,059
5,481
5,904
21. Lancashire
3.40%
6,371
6,902
7,434
22.Leicestershire
1.40%
2,623
2,842
3,061
23. Lincolnshire
0.60%
1,124
1,218
1,312
Schedule F, Service Level Requirements
26
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Monthly Mark-to-Print Searches (MPmax)
Bureaux
Distribution
2004
2005
2006
24. Merseyside
3.30%
6,183
6,699
7,215
25. Metropolitan
17.10%
32,041
34,715
37,389
26. Norfolk
1.30%
2,436
2,639
2,842
27. North Wales
0.90%
1,686
1,827
1,968
28. North Yorkshire
0.90%
1,686
1,827
1,968
29. Northhamptonshire
1.80%
3,373
3,654
3,936
30. Northumbria
1.90%
3,560
3,857
4,154
31. Nottinghamshire
1.30%
2,436
2,639
2,842
32. South Wales
1.70%
3,185
3,451
3,717
33. South Yorkshire
1.10%
2,061
2,233
2,405
34. Staffordshire
2.30%
4,310
4,669
5,029
35. Suffolk
0.40%
749
812
875
36. Surrey
1.10%
2,061
2,233
2,405
37. Sussex
2.00%
3,747
4,060
4,373
38. Thames Valley
3.30%
6,183
6,699
7,215
39. Warwickshire
0.60%
1,124
1,218
1,312
40. West Mercia
1.00%
1,874
2,030
2,186
41. West Midlands
4.30%
8,057
8,729
9,402
42. West Yorkshire
6.50%
12,179
13,196
14,212
43. Wiltshire
0.70%
1,312
1,421
1,531
44. Aberdeen
0
0
0
Included in Glasgow
45. Dundee
0
0
0
below
46. Edinburgh
0
0
0
47. Glasgow
6.50%
12,179
13,196
14,212
48. British Transport Police
0.30%
562
609
656
49. HM Customs & Excise
0.10%
187
203
219
50. National Identification
1.30%
2,436
2,639
2,842
Service
51. National Crime Squad
0
0
0
Schedule F, Service Level Requirements
27
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Monthly Mark-to-Print Searches (MPmax)
Bureaux
Distribution
2004
2005
2006
52. Specialist Crimes
0
0
0
Bureau
Total (Upper Bound)
100.00%
187,374
203,011
218,648
Figure 2.2-6 MPmax per Bureau
To account for minor variations from the DOR specified distribution, starting in year 2007,
MPmax will be based on the actual distribution of the Mark to Print searches between
bureaux measured during a 12 month window in the year prior to the preceding year. For each
bureau, the average monthly Mark to Print search load will be calculated. These averages will
then be used to compute the actual distribution between bureaux subject to the constraint that
the total Mark to Print search load is as specified by DOR Tables 2.17 and 2.18. Due to the
changing nature of MPmax and TPmax, CIP will not be applicable to the Throughput service
level criteria.
Note that in Figures 2.2-5 and 2.2-6, all Scottish workloads are grouped under Glasgow. Prior
to the IOC of the Scottish Bureaux, workload allocation to individual bureaux will be defined
and agreed.
2.2.4.2 Throughput Buffer Zone
When a Bureau exceeds its allocation of the workload as defined by these parameters,
they can continue to process ten prints and submit mark-to-print searches. However,
to compensate for potential degradation in system performance, the throughput
metrics under these circumstances can exceed 100% and the score greater than the
target score of 100. To allow for potential variation in bureaux workloads from those
specified in the DOR, changes will be made to the throughput metric calculation
equations (equations 2.2-1 of section 2.2.2 and 2.2-2 of section 2.2.3). These changes
will be effective starting at TOR and are shown below.
TPc
Equation 2.2-3:
TPt = MIN(
×
TPa 1
, . xTP max)
100
1
( ni
M +
i
M +
r
M + Mpni +
pi
M )
Equation 2.2-4:
t
M = MIN(
×
MPa 1
, . xMP max)
100
1
+ pr
M
Note that the “1.1” multiplier for TPmax and MPmax effectively provides a 10%
buffer zone above the maximum Ten Print and Mark workload allocated to each
Bureau to account for potential variation in workload distribution.
2.2.4.3 Scottish Bureaux
Since the above tables include the workload allocations to the Scottish Bureaux, all
that is needed when the Scottish Bureaux become operational is to start the
measurement and scoring using the same throughput metrics defined for existing
Bureaux.
2.2.4.4 Palm Searches
Schedule F, Service Level Requirements
28
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Palm marks are part of the total search load specified in Table 2.17 of DOR Annex A
from which the MPmax parameter for each Bureau is calculated. Inclusion of the
palm marks in throughput measurement of mark-to-print searches when the capability
becomes available therefore will not require any changes to the metric.
2.3
Search Performance
System Performance will be measured by the search response time delivered to each Bureau.
Response time will be measured at the Bureau Segment. The time will start when the search
request is submitted by the user and the time will end when the requested transaction is
available to the user. This will include image transmission time.
System performance is dependent on the distribution and type of search requests. Deviations
outside the ranges specified in this SLR, stated as MAXIMUM, will influence performance
levels. The major influences are:
(a)
Variation in inter-transaction arrival rate within a given hour. System performance is
predicated on a statistical distribution of transactions within the hour. If more
transactions are received than expected in a specified period of time, the system
response will be slower than expected as the system attempts to process and clear the
larger than expected queues of transactions.
(b)
Variation in the mix between Ten Print and Mark submissions. Mark processing
consumes more system resources than Ten Print processing. If more Marks are
submitted than expected during a given time period, system performance will
decrease from nominal.
(c)
Variation in the distribution of transaction submissions across the bureaux. A larger
than expected number of submissions from a single bureau could exceed the inherent
capability of the Bureau Front End System in terms of the number of available
workstations, number of fingerprint officers, and WAN bandwidth. All of these
effects would contribute to a decrease in response performance.
(d)
Changes to the nominal search parameters. The search parameters expected over an
average set of submissions have been established (see Section 5 of this SLR) which
will result in a set of averaged response timings. In certain situations, it may be
desirable for a Bureau to change the nominal values of search parameters in a manner
that increases the number of candidate respondents. This would slow system
response by causing additional image retrieval, transmitting more data over the
WAN, and increasing the decompression processing time in the bureau.
(e)
Changes to the nominal assumptions of how work at the bureaux is distributed
amongst the system resources (primarily workstations). The design of the system
offers great flexibility in configuring and managing the bureau workload. However,
nominal performance is based on the assumption of local work procedures that make
efficient use of the system resources. Local work procedures may be implemented
that do not fully utilise system resources, and thus have an adverse effect on response
time for those processes performed at the local bureau.
System Performance will be measured for each of the four search types: Print-to-Print, Print-
to-Mark, Mark-to-Print, and Mark-to-Mark. Print-to-Mark searches are performed on two
National repositories, the National Unidentified Mark Collection and the Serious Crime
Cache. These searches are separately measured; the Serious Crime Cache measurements are
recorded but are unweighted. In addition, within each search type, system performance will be
presented for each of the three available search priorities: Urgent, Fast, and Normal. Serious
Schedule F, Service Level Requirements
29
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Crime Cache, Operational Response, and External Transaction Bureau searches are all
submitted as Normal priority. IND P-M searches have a priority of Low and these are
presented in the P-M performance table.
The expected distribution of search scope by search priority for each search type is defined in
Section 5 of this SLR. In the equations below, a premium will be placed on searches
submitted in excess of the maximum allowed for the search priority.
2.3.1
Print-to-Print Performance
Equation 2.3-1 will be used to compute the PRINT-TO-PRINT PERFORMANCE
value for each of the search priorities x within a Print-to-Print search type. Equation
2.3-1 will be applied once for each of the search priorities and once for Operational
Response searches.
c
X
Equation 2.3-1:
P − x
P =
×100
a
X
where
(a)
P-Px is the PRINT-TO-PRINT PERFORMANCE value for each search
priority x: Urgent, Fast, and Normal or Operational Response.
(b)
Xc is the number of searches, of the specified priority x, that complete within
the Target Response Times defined in Figure 2.3-1.
Search Type
Category
Priority
Target Response
Time (Mins)
Print-to-Print
Central
Urgent
3
Central
Fast
5
Central
Normal
7
ORD
Normal
7
Figure 2.3-1 Print-to-Print Target Response Times
(c)
Xa is the actual number of searches, of the specified priority x, that were
submitted by the bureau during the accounting period. If Xa is zero during an
accounting period, all expressions that include Xa will be set to zero in order
to avoid undefined mathematical expressions.
The PRINT-TO-PRINT PERFORMANCE value for each search priority and
Operational Response searches for each Bureau will be computed using Equation 2.3-
1. These values will be presented for information and service analysis purposes.
Then the OVERALL PRINT-TO-PRINT PERFORMANCE value will be calculated
and entered in Figure 2.3-2. This will be calculated by dividing the total number of
Print-to-Print searches that complete within the Response Times defined in Figure
2.3-1, by the total number of Print-to-Print searches submitted during the accounting
period.
Schedule F, Service Level Requirements
30
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Overall
Print-to-
Print-to-Print
Print-to-
Print-to- Print-to-
Print-to-Print
Print
Bureau
Urgent Perf.
Print ORD
Print
Print
Fast Perf. Value Normal Perf.
Value
Perf. Value
Perf.
SCORE
Value
Value
Avon + Somerset
Bedfordshire
British Transport
Police
Cambridgeshire
Cheshire
City of London
Cleveland
Cumbria
Customs and Excise
Derbyshire
Devon + Cornwall
Dorset
Durham
Dyfed Powys
Essex
Gloucestershire
Greater Manchester
Gwent
Hampshire
Hertfordshire
Humberside
Kent
Lancashire
Leicestershire
Lincolnshire
Schedule F, Service Level Requirements
31
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Merseyside
Metropolitan
N.I.S./Residual
Bureau
Norfolk
North Wales
North Yorkshire
Northamptonshire
Northumbria
Nottinghamshire
South Wales
South Yorkshire
Staffordshire
Suffolk
Surrey
Sussex
Thames Valley
Warwickshire
West Mercia
West Midlands
West Yorkshire
Wiltshire
Ext. Transaction
Bureau
Figure 2.3-2 Print-to-Print Performance Table
In the event that no applicable print-to-print searches are processed by the site in the
reporting period, a fixed performance value of 90 (corresponding to a SCORE of 100)
shall be applied to the OVERALL PRINT-TO-PRINT PERFORMANCE. Once the
OVERALL PRINT-TO-PRINT PERFORMANCE value has been computed for each
Bureau, the SCORE for each Bureau will be selected from the Scoring Table
presented as Figure 2.3-3. The absence of searches of a particular priority has no
effect on the OVERALL SCORE, but will be indicated in the Table with “N/A”.
Schedule F, Service Level Requirements
32
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Print-to-Print
Print-to-Print
SCORE
SCORE
Performance
Performance
100
120
79
89
99.5
119
78
88
99
118
77
87
98.5
117
76
86
98
116
75
85
97.5
115
74
84
97
114
73
83
96.5
113
72
82
96
112
71
81
95.5
111
70
80
95
110
69
79
94.5
109
68
78
94
108
67
77
93.5
107
66
76
93
106
65
75
92.5
105
64
74
92
104
63
73
91.5
103
62
72
91
102
61
71
90.5
101
60
70
90
100
59
69
89
99
58
68
88
98
57
67
87
97
56
66
86
96
55
65
Schedule F, Service Level Requirements
33
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
85
95
54
64
84
94
53
63
83
93
52
62
82
92
51
61
81
91
≤51
60
80
90
Figure 2.3-3 Print-to-Print Performance Scoring Table
Schedule F, Service Level Requirements
34
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
The OVERALL PRINT-TO-PRINT PERFORMANCE SCORE will be used for SLR
reporting as shown in Figure 4.1-1. Print-to-Print Performance will be reported on an
individual Bureau basis.
2.3.2
Print-to-Mark Performance
Equation 2.3-2 will be used to compute the PRINT-TO-MARK PERFORMANCE
value for each of the three search priorities x within a Print-to-Mark search type.
Equation 2.3-2 will be applied once for each of the three search priorities, and once
for Operational Response searches. In addition, a priority of LOW with a response
time of 24 hours supports P-M searches from IND. All P-M searches from IND will
be submitted with a LOW priority. Equation 2.3-2 will be applied a fifth time for
these IND searches.
c
X
Equation 2.3-2:
P −
x
M =
×100
a
X
where
(a)
P-Mx is the PRINT-TO-MARK PERFORMANCE value for each search
priority x: Urgent, Fast, Normal, and Low or Operational Response.
(b)
Xc is the number of searches, of the specified search priority x, that complete
within the Target Response Times as defined in Figure 2.3-4.
Target Response
Search Type
Category
Priority
Times
Print-to-Mark
Central
Urgent
12 minutes
Central
Fast
18 minutes
Central
Normal
50 minutes
Central
Low
24 hours
ORD
Normal
12 minutes
Figure 2.3-4 Print-to-Mark Target Response Times
(c)
Xa is the actual number of searches, of the specified search priority x, that
were submitted by the bureau during the accounting period. If Xa is zero
during an accounting period, all expressions that include Xa will be set to zero
so as to avoid undefined mathematical expressions.
The PRINT-TO-MARK PERFORMANCE value for each search priority for each
Bureau will be computed using Equation 2.3-2. These values will be presented for
information and service analysis purposes. The British Transport Police (BTP),
Customs and Excise Service (CAE), and ETB Bureaux may perform all national
searches rather than the mix of local/regional/national searches indicated in Section 5.
For these three Bureaux, the Print-to-Mark Performance value will be calculated for
each search priority using Equation 2.3-3.
Schedule F, Service Level Requirements
35
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
c
X
Equation 2.3-3:
P −
x
M =
×100 +10
a
X
In the event that Equation 2.3-3 results in a Performance value greater than 100, the
Performance value shall be set to 100.
Then the OVERALL PRINT-TO-MARK PERFORMANCE value will be calculated
and entered in Figure 2.3-5. This will be calculated by dividing the total number of
Print-to-Mark searches of all search priorities and Operational Response searches that
complete within the Response Times as defined in Figure 2.3-6, by the total number
of Print-to-Mark searches submitted during the accounting period. For BTP, CAE,
and ETB, the Overall Print-to-Mark Performance value will be calculated by dividing
the total number of Print-to-Mark searches completed within the required Response
Times by the total number of Print-to-Mark searches submitted, multiplying by 100,
and then adding 10.
Bureau
Print-to-
Print-to- Print-
Print- Print-to-
Overall
Print-to-
Mark
Mark
to-
to-
Mark
Print-to-
Mark
Urgent
Fast
Mark
Mark
ORD
Mark Perf.
SCORE
Perf. Value
Perf.
Normal Low
Perf.
Value
Value
Perf.
Perf.
Value
Value
Value
Avon + Somerset
Bedfordshire
British Transport
Police
Cambridgeshire
Cheshire
City of London
Cleveland
Cumbria
Customs and Excise
Derbyshire
Devon + Cornwall
Dorset
Durham
Dyfed Powys
Essex
Schedule F, Service Level Requirements
36
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Bureau
Print-to-
Print-to- Print-
Print- Print-to-
Overall
Print-to-
Mark
Mark
to-
to-
Mark
Print-to-
Mark
Urgent
Fast
Mark
Mark
ORD
Mark Perf.
SCORE
Perf. Value
Perf.
Normal Low
Perf.
Value
Value
Perf.
Perf.
Value
Value
Value
Gloucestershire
Greater Manchester
Gwent
Hampshire
Hertfordshire
Humberside
Kent
Lancashire
Leicestershire
Lincolnshire
Merseyside
Metropolitan
N.I.S./Residual
Bureau
Norfolk
North Wales
North Yorkshire
Northamptonshire
Northumbria
Nottinghamshire
South Wales
South Yorkshire
Staffordshire
Suffolk
Schedule F, Service Level Requirements
37
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Bureau
Print-to-
Print-to- Print-
Print- Print-to-
Overall
Print-to-
Mark
Mark
to-
to-
Mark
Print-to-
Mark
Urgent
Fast
Mark
Mark
ORD
Mark Perf.
SCORE
Perf. Value
Perf.
Normal Low
Perf.
Value
Value
Perf.
Perf.
Value
Value
Value
Surrey
Sussex
Thames Valley
Warwickshire
West Mercia
West Midlands
West Yorkshire
Wiltshire
Ext. Transaction
Bureau
Figure 2.3-5 Print-to-Mark Performance Table
In the event that no applicable print-to-mark searches are processed by the site in the reporting period,
a fixed performance value of 90 (corresponding to a SCORE of 100) shall be applied to the
OVERALL PRINT-TO-MARK PERFORMANCE. This shall also apply to the BTP, CAE, and ETB
bureaux. Once the OVERALL PRINT-TO-MARK PERFORMANCE value has been computed for
each Bureau, the SCORE for each Bureau will be selected from the Scoring Table presented as Figure
2.3-1. The absence of searches of a particular priority has no effect on the OVERALL SCORE, but
will be indicated in the Table with “N/A”.
Print-to-Mark
Print-to-Mark
Print-to-Mark
SCORE
SCORE
SCORE
Performance
Performance
Performance
100
120
89
99
68
78
99.5
119
88
98
67
77
99
118
87
97
66
76
98.5
117
86
96
65
75
98
116
85
95
64
74
97.5
115
84
94
63
73
97
114
83
93
62
72
96.5
113
82
92
61
71
Schedule F, Service Level Requirements
38
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
96
112
81
91
60
70
95.5
111
80
90
59
69
95
110
79
89
58
68
94.5
109
78
88
57
67
94
108
77
87
56
66
93.5
107
76
86
55
65
93
106
75
85
54
64
92.5
105
74
84
53
63
92
104
73
83
52
62
91.5
103
72
82
51
61
91
102
71
81
<51
60
90.5
101
70
80
90
100
69
79
Figure 2.3-6 Print-to-Mark Performance Scoring Table
Schedule F, Service Level Requirements
39
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
The PRINT-TO-MARK PERFORMANCE SCORE will be used for SLR reporting as shown
in Figure 4.1-1. Print-to-Mark Performance will be reported on an individual Bureau basis.
2.3.3
Print-to-Mark Serious Crime Cache Performance
Print to Mark Serious Crime Cache performance is reported but does not contribute to
the SLR Scoring from the milestone TOR – NAFIS onwards.
Equation 2.3-4 will be used to compute the PRINT-TO-MARK PERFORMANCE
value for the Print-to-Mark Serious Crime Cache search type.
Equation
P − M =
c
X
x
×100
2.3-4:
a
X
where
(a)
P-Mx
is
the
PRINT-TO-MARK
SERIOUS
CRIME
CACHE
PERFORMANCE value.
(b)
Xc is the number of searches that complete within the Target Response Times
as defined in Figure 2.3-7. All Print-to-Mark Serious Crime Cache searches
are submitted as Normal priority.
(c)
Xa is the actual number of searches that were submitted by the bureau during
the accounting period. If Xa is zero during an accounting period, all
expressions that include Xa will be set to zero so as to avoid undefined
mathematical expressions.
Search Type
Priority
Target Response Times
Print-to-Mark Serious
Normal
8 minutes
Crime Cache
Figure 2.3-7 Print-to-Mark Serious Crime Cache Target Response Time
The PRINT-TO-MARK SERIOUS CRIME CACHE PERFORMANCE value for each Bureau will be
computed using Equation 2.3-4. These values will be presented for information and service analysis
purposes.
Then the OVERALL PRINT-TO-MARK SERIOUS CRIME CACHE PERFORMANCE value will
be calculated and entered in Figure 2.3-8. This will be calculated by dividing the total number of
Print-to-Mark Serious Crime Cache searches that complete within the Response Times as defined in
Figure 2.3-8, by the total number of Print-to-Mark Serious Crime Cache searches submitted during
the accounting period.
Schedule F, Service Level Requirements
40
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Bureau
Print-to-Mark Serious
Bureau
Print-to-Mark Serious
Crime Cache Normal
Crime Cache Normal
Perf. Value
Perf. Value
Avon + Somerset
Metropolitan
Bedfordshire
-
British Transport
N.I.S./Residual
Police
Bureau
Cambridgeshire
Norfolk
Cheshire
North Wales
City of London
North Yorkshire
Cleveland
Northamptonshire
Cumbria
Northumbria
Customs and Excise
Nottinghamshire
Derbyshire
South Wales
Devon + Cornwall
South Yorkshire
Dorset
Staffordshire
Durham
Suffolk
Dyfed Powys
Surrey
Essex
Sussex
Gloucestershire
Thames Valley
Greater Manchester
Warwickshire
Gwent
West Mercia
Hampshire
West Midlands
Hertfordshire
West Yorkshire
Humberside
Wiltshire
Kent
Ext. Transaction
Bureau
Lancashire
Figure 2.3-8 Print-to-Mark Performance
Leicestershire
Serious Crime Cache Table
Lincolnshire
Merseyside
Schedule F, Service Level Requirements
41
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
In the event that no applicable print-to-mark serious crime cache searches are processed by the site in
the reporting period, a fixed performance value of 90 shall be applied to the OVERALL PRINT-TO-
MARK SERIOUS CRIME CACHE PERFORMANCE.
2.3.4
Mark-to-Print Performance
Equation 2.3-5 will be used to compute the MARK-TO-PRINT PERFORMANCE
value for each of the three search priorities x within a Mark-to-Print search type.
Equation 2.3-5 will be applied four times, once for each of the three search priorities,
and once for Operational Response searches.
c
X
Equation 2.3-5:
M − x
P =
×100
a
X
where
(a)
M-Px is the MARK-TO-PRINT PERFORMANCE value for each search
priority x: Urgent, Fast, and Normal or Operational Response.
(b)
Xc is the number of searches, of the specified search priority x, that complete
within the Target Response Times as defined in Figure 2.3-9.
Target Response
Search Type
Category
Priority
Times
Mark-to-Print
Central
Urgent
6 minutes
Central
Fast
10 minutes
Central
Normal
20 minutes
ORD
Normal
10 minutes
Figure 2.3-9 Mark-to-Print Target Response Times
(c)
Xa is the actual number of searches, of the specified search priority x, that
were submitted by the bureau during the accounting period. If Xa is zero
during an accounting period, all expressions that include Xa will be set to zero
so as to avoid undefined mathematical expressions.
The MARK-TO-PRINT PERFORMANCE value for each search priority for each
Bureau will be computed using Equation 2.3-5. These values will be presented for
information and service analysis purposes. The British Transport Police (BTP) and
Customs and Excise (CAE) Bureaux may perform all national searches rather than the
mix of local/regional/national searches indicated in Section 5. For these two
Bureaux, the Mark-to-Print Performance value will be calculated for each search
priority using Equation 2.3-6.
c
X
Equation 2.3-6:
M − x
P =
×100 +10
a
X
In the event that Equation 2.3-6 results in a Performance value greater than 100, the
Performance value shall be set to 100.
Schedule F, Service Level Requirements
42
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Then the OVERALL MARK-TO-PRINT PERFORMANCE value will be calculated
and entered in Figure 2.3-10. This will be calculated by dividing the total number of
Mark-to-Print searches of all search priorities and Operational Response searches that
complete within the Response Times defined in Figure 2.3-9, by the total number of
Mark-to-Print searches submitted during the accounting period. For BTP & CAE, the
Overall Mark-to-Print Performance value will be calculated by dividing the total
number of Mark-to-Print searches completed within the required Response Times by
the total number of Mark-to-Print searches submitted, multiplying by 100, and then
adding 10.
Overall
Mark-to-
Mark-to-Print Mark-to-Print
Mark-to-
Mark-
Mark-
Print
Bureau
Urgent Perf.
Fast Perf.
Print ORD to-Print to-Print
Normal Perf.
Value
Value
Perf. Value
Perf.
SCORE
Value
Value
Avon + Somerset
Bedfordshire
British Transport
Police
Cambridgeshire
Cheshire
City of London
Cleveland
Cumbria
Customs and Excise
Derbyshire
Devon + Cornwall
Dorset
Durham
Dyfed Powys
Essex
Gloucestershire
Greater Manchester
Gwent
Schedule F, Service Level Requirements
43
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Overall
Mark-to-
Mark-to-Print Mark-to-Print
Mark-to-
Mark-
Mark-
Print
Bureau
Urgent Perf.
Fast Perf.
Print ORD to-Print to-Print
Normal Perf.
Value
Value
Perf. Value
Perf.
SCORE
Value
Value
Hampshire
Hertfordshire
Humberside
Kent
Lancashire
Leicestershire
Lincolnshire
Merseyside
Metropolitan
N.I.S./Residual
Bureau
Norfolk
North Wales
North Yorkshire
Northamptonshire
Northumbria
Nottinghamshire
South Wales
South Yorkshire
Staffordshire
Suffolk
Surrey
Sussex
Thames Valley
Schedule F, Service Level Requirements
44
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Overall
Mark-to-
Mark-to-Print Mark-to-Print
Mark-to-
Mark-
Mark-
Print
Bureau
Urgent Perf.
Fast Perf.
Print ORD to-Print to-Print
Normal Perf.
Value
Value
Perf. Value
Perf.
SCORE
Value
Value
Warwickshire
West Mercia
West Midlands
West Yorkshire
Wiltshire
Ext. Transaction
Bureau
Figure 2.3-10 Mark-to-Print Performance Table
Schedule F, Service Level Requirements
45
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
In the event that no applicable mark-to-print searches are processed by the site in the
reporting period, a fixed performance value of 90 (corresponding to a SCORE of 100)
shall be applied to OVERALL MARK-TO-PRINT PERFORMANCE. This shall also
apply to the BTP and CAE Bureaux IND is not expected to submit any Mark-to-Print
searches and the ETB SCORE will be set to a performance value of 90
(corresponding to a SCORE of 100). Once the OVERALL MARK-TO-PRINT
PERFORMANCE value has been computed for each Bureau, the SCORE for each
Bureau will be selected from the Scoring Table presented as Figure 2.3-11. The
absence of searches of a particular priority has no effect on the OVERALL SCORE,
but will be indicated in the Table with “N/A”.
Mark-to-Print
Mark-to-Print
Mark-to-Print
SCORE
SCORE
SCORE
Performance
Performance
Performance
100
120
89
99
68
78
99.5
119
88
98
67
77
99
118
87
97
66
76
98.5
117
86
96
65
75
98
116
85
95
64
74
97.5
115
84
94
63
73
97
114
83
93
62
72
96.5
113
82
92
61
71
96
112
81
91
60
70
95.5
111
80
90
59
69
95
110
79
89
58
68
94.5
109
78
88
57
67
94
108
77
87
56
66
93.5
107
76
86
55
65
93
106
75
85
54
64
92.5
105
74
84
53
63
92
104
73
83
52
62
91.5
103
72
82
51
61
91
102
71
81
(51
60
90.5
101
70
80
90
100
69
79
Figure 2.3-11 Mark-to-Print Performance Scoring Table
Schedule F, Service Level Requirements
46
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
The MARK-TO-PRINT PERFORMANCE SCORE will be used for SLR reporting as
shown in Figure 4.1-1. Mark-to-Print Performance will be reported on an individual
Bureau basis.
2.3.5
Mark-to-Mark Performance
Equation 2.3-7 will be used to compute the MARK-TO-MARK PERFORMANCE
value for each of the three search priorities x within a Mark-to-Mark search type.
Equation 2.3-7 will be applied four times, once for each of the three search priorities,
and once for Operational Response searches.
c
X
Equation 2.3-7:
M −
x
M =
×100
a
X
where
(a)
M-Mx is the MARK-TO-MARK PERFORMANCE value for each search
priority x: Urgent, Fast, and Normal or Operational Response.
(b)
Xc, is the number of searches, of the specified search priority x, that complete
within the Target Response Times defined in Figure 2.3-12.
Target response
Search Type
Category
Priority
times
Mark-to-Mark
Central
Urgent
3 minutes
Central
Fast
7 minutes
Central
Normal
20 minutes
ORD
Normal
10 minutes
Figure 2.3-12 Mark-to-Mark Target Response Times
(c)
Xa is the actual number of searches, of the specified search priority x, that
were submitted by the bureau during the accounting period. If Xa is zero
during an accounting period, all expressions that include Xa will be set to zero
so as to avoid undefined mathematical expressions.
The MARK-TO-MARK PERFORMANCE value for each search priority for each
Bureau will be computed using Equation 2.3-7. These values will be presented for
information and service analysis purposes. The British Transport Police (BTP) and
Customs and Excise (CAE) Bureaux may perform all national searches rather than the
mix of local/regional/national searches indicated in Section 5. For these two
Bureaux, the Mark-to-Mark Performance value will be calculated for each search
priority using Equation .2.3.8.
c
X
Equation 2.3-8:
M −
x
M =
×100 +10
a
X
In the event that Equation 2.3-8 results in a Performance value greater than 100, the
Performance value shall be set to 100.
Schedule F, Service Level Requirements
47
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Then the OVERALL MARK-TO-MARK PERFORMANCE value will be calculated
and entered in Figure 2.3-13. This will be calculated by dividing the total number of
Mark-to-Mark searches of all search priorities and Operational Response searches
that complete within the Response Times as defined in Figure 2.3-12, by the total
number of Mark-to-Mark searches submitted during the accounting period. For BTP
& CE the Overall Mark-to-Mark Performance value will be calculated by dividing the
total number of Mark-to-Mark searches completed within the required Response
Times by the total number of Mark-to-Mark searches submitted, multiplying by 100,
and then adding 10.
Overall
Mark-to-
Mark-to-
Mark-to-
Mark-
Marl-to-
Mark-to-Mark
Mark
Mark
Bureau
Mark Urgent
to-Mark Mark
Fast Perf.
Normal Perf. ORD Perf.
Perf. Value
Perf.
SCORE
Value
Value
Value
Value
Avon + Somerset
Bedfordshire
British Transport
Police
Cambridgeshire
Cheshire
City of London
Cleveland
Cumbria
Customs and Excise
Derbyshire
Devon + Cornwall
Dorset
Durham
Dyfed Powys
Essex
Gloucestershire
Greater Manchester
Gwent
Schedule F, Service Level Requirements
48
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Overall
Mark-to-
Mark-to-
Mark-to-
Mark-
Marl-to-
Mark-to-Mark
Mark
Mark
Bureau
Mark Urgent
to-Mark Mark
Fast Perf.
Normal Perf. ORD Perf.
Perf. Value
Perf.
SCORE
Value
Value
Value
Value
Hampshire
Hertfordshire
Humberside
Kent
Lancashire
Leicestershire
Lincolnshire
Merseyside
Metropolitan
N.I.S./Residual
Bureau
Norfolk
North Wales
North Yorkshire
Northamptonshire
Northumbria
Nottinghamshire
South Wales
South Yorkshire
Staffordshire
Suffolk
Surrey
Sussex
Thames Valley
Schedule F, Service Level Requirements
49
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Overall
Mark-to-
Mark-to-
Mark-to-
Mark-
Marl-to-
Mark-to-Mark
Mark
Mark
Bureau
Mark Urgent
to-Mark Mark
Fast Perf.
Normal Perf. ORD Perf.
Perf. Value
Perf.
SCORE
Value
Value
Value
Value
Warwickshire
West Mercia
West Midlands
West Yorkshire
Wiltshire
Ext. Transaction
Bureau
Figure 2.3-13 Mark-to-Mark Performance Table
In the event that no applicable mark-to-mark searches are processed by the site in the
reporting period, a fixed performance value of 90 (corresponding to a SCORE of 100)
shall be applied to OVERALL MARK-TO-MARK PERFORMANCE VALUE. This
shall also apply to the BTP and CAE Bureaux. IND is not expected to submit any
Mark-to-Mark searches and the ETB SCORE will be set to a performance value of 90
(corresponding to a SCORE of 100). Once the OVERALL MARK-TO-MARK
PERFORMANCE value has been computed for each Bureau, the SCORE for each
Bureau will be selected from the Scoring Table presented as Figure 2.3-14. The
absence of searches of a particular priority has no effect on the OVERALL SCORE,
but will be indicated in the Table with “N/A”.
Mark-to-Mark
Mark-to-Mark
Mark-to-Mark
SCORE
SCORE
SCORE
Performance
Performance
Performance
100
120
89
99
68
78
99.5
119
88
98
67
77
99
118
87
97
66
76
98.5
117
86
96
65
75
98
116
85
95
64
74
97.5
115
84
94
63
73
97
114
83
93
62
72
96.5
113
82
92
61
71
96
112
81
91
60
70
Schedule F, Service Level Requirements
50
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
95.5
111
80
90
59
69
95
110
79
89
58
68
94.5
109
78
88
57
67
94
108
77
87
56
66
93.5
107
76
86
55
65
93
106
75
85
54
64
92.5
105
74
84
53
63
92
104
73
83
52
62
91.5
103
72
82
51
61
91
102
71
81
<51
60
90.5
101
70
80
90
100
69
79
Figure 2.3-14 Mark-to-Mark Performance Scoring Table
The MARK-TO-MARK PERFORMANCE SCORE will be used for SLR reporting as
shown in Figure 4.1-1. Mark-to-Mark Performance will be reported on an individual
Bureau basis.
2.3.6
Quoted CRO Image Retrieval
Equation 2.3-9 will be used to compute the QUOTED CRO IMAGE RETRIEVAL
PERFORMANCE value. Quoted CRO Image Retrievals are not like AFR searches in
that no AFR Search of the Central Data Bases is required. Instead, the user’s request
for image data, as requested as the result of a receipt of a Quoted CRO Ten Print
form, initiates the transaction. The QUOTED CRO IMAGE RETRIEVAL
PERFORMANCE equation shown in Equation 2.3-9 does not include provisions for
retrievals based on priority.
Within IDENT1, the Quoted CRO Image Retrieval Transaction will start when the
user “accepts” the transaction at the end of Demographic Data Check (to include the
Quoted CRO entered into the screen by the user). The transaction will end when the
image data is available on the Bureau Server. At this point, the user may view the
data.
Schedule F, Service Level Requirements
51
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Equation 2.3-9 will be applied once.
c
X
Equation 2.3-9:
QUOTED =
×100
a
X
where
(a)
QUOTED is the QUOTED CRO IMAGE RETRIEVAL PERFORMANCE
value.
(b)
Xc, is the number of Quoted CRO Image Retrieval transactions that complete
within the Target Response Times as defined in Figure 2.3-15.
Target Response Times
Search Type
(seconds)
Quoted CRO
60
Figure 2.3-15 Quoted CRO Image Retrieval Target Response Times
(c)
Xa is the actual number of Quoted CRO Image Retrieval transactions that
were submitted by the bureau during the accounting period. If Xa is zero
during an accounting period, all expressions that include Xa will be set to zero
so as to avoid undefined mathematical expressions.
The QUOTED CRO IMAGE RETRIEVAL PERFORMANCE value for each Bureau
will be computed using Equation 2.3-9. This value will be placed in Figure 2.3-16.
QUOTED CRO IMAGE
QUOTED CRO IMAGE
Bureau
RETRIEVAL
RETRIEVAL
PERFORMANCE VALUE
PERFORMANCE SCORE
Avon + Somerset
Bedfordshire
British Transport Police
Cambridgeshire
Cheshire
City of London
Cleveland
Customs and Excise
Cumbria
Derbyshire
Devon + Cornwall
Schedule F, Service Level Requirements
52
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
QUOTED CRO IMAGE
QUOTED CRO IMAGE
Bureau
RETRIEVAL
RETRIEVAL
PERFORMANCE VALUE
PERFORMANCE SCORE
Dorset
Durham
Dyfed Powys
Essex
Gloucestershire
Greater Manchester
Gwent
Hampshire
Hertfordshire
Humberside
Kent
Lancashire
Leicestershire
Lincolnshire
Merseyside
Metropolitan
N.I.S./Residual Bureau
Norfolk
North Wales
North Yorkshire
Northamptonshire
Northumbria
Nottinghamshire
South Wales
South Yorkshire
Schedule F, Service Level Requirements
53
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
QUOTED CRO IMAGE
QUOTED CRO IMAGE
Bureau
RETRIEVAL
RETRIEVAL
PERFORMANCE VALUE
PERFORMANCE SCORE
Staffordshire
Suffolk
Surrey
Sussex
Thames Valley
Warwickshire
West Mercia
West Midlands
West Yorkshire
Wiltshire
Ext. Transaction Bureau
Figure 2.3-16 Quoted CRO Performance Table
In the event that no applicable quoted CRO image retrievals are processed by the site
in the reporting period, a fixed SCORE of 100 shall be applied to QUOTED CRO
IMAGE RETRIEVAL PERFORMANCE VALUE. This shall also apply to the BTP,
CAE and ETB Bureaux. Once the QUOTED CRO IMAGE RETRIEVAL
PERFORMANCE value has been computed for each Bureau, the SCORE for each
Bureau will be selected from the Scoring Table presented as Figure 2.3-17.
Schedule F, Service Level Requirements
54
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
QUOTED
QUOTED
QUOTED
SCORE
SCORE
SCORE
Performance
Performance
Performance
100
120
89
99
68
78
99.5
119
88
98
67
77
99
118
87
97
66
76
98.5
117
86
96
65
75
98
116
85
95
64
74
97.5
115
84
94
63
73
97
114
83
93
62
72
96.5
113
82
92
61
71
96
112
81
91
60
70
95.5
111
80
90
59
69
95
110
79
89
58
68
94.5
109
78
88
57
67
94
108
77
87
56
66
93.5
107
76
86
55
65
93
106
75
85
54
64
92.5
105
74
84
53
63
92
104
73
83
52
62
91.5
103
72
82
51
61
91
102
71
81
<51
60
90.5
101
70
80
90
100
69
79
Figure 2.3-17 Quoted CRO Image Retrieval Performance Scoring Table
The QUOTED CRO IMAGE RETRIEVAL PERFORMANCE SCORE will be used
for SLR reporting as shown in Figure 4.1-1. Quoted CRO Image Retrieval
Performance will be reported on an individual Bureau basis.
2.3.7
Future Changes for Search Response Time
There are several changes for the response time service level criteria after TOR.
Schedule F, Service Level Requirements
55
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
These changes are described below. Response time measurements for marks (using
existing metrics) will include palm marks when the palm search capability becomes
operational.
2.3.7.1 Improved Target Response Times
In addition to the significantly improved TOR target response times (vs NAFIS)
specified in the previous sections, faster target response times for most search types
will be implemented for FOC and FOC+3 to provide continuing incremental
improvements. These 90 percentile target values are summarised in Figure 2.3-18.
Note that due to the fast response times, the Fast priority has been eliminated.
In addition, a new search priority category for searches that can take advantage of the
off-peak processing capacity may be introduced if agreed by the Authority and
Contractor.
90% Target Response
Database
Time (Minutes)
Search Type
Category
Priority
Penetration
FOC
FOC+3
Print-to-Print
100%
Central
Urgent
3
2
Central
Normal
4
3
ORD
Normal
4
3
Print-to-Mark
100%
Central
Urgent
30
22
Central
Normal
40
30
ORD
Normal
10
8
Finger Mark-to-Print
100% for 15% of
Central
Urgent
5
4
searches and 30%
for 85% of searches
Central
Normal
15
11
ORD
Normal
8
6
Palm Mark-to-Print
100% for 15% of
Central
Urgent
15
12
searches and 30%
for 85% of searches
≤ 30%
Central
Normal
15
12
ORD
Normal
8
6
> 30% and ≤ 100%
Central
Normal
45
36
for up to 15% of
searches
ORD
Normal
15
10
Schedule F, Service Level Requirements
56
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Mark-to-Mark
100%
Central
Urgent
5
4
Central
Normal
10
8
ORD
Normal
5
4
Figure 2.3-18 90% Target Response Times at FOC and FOC+3
Note that faster response times for searches against the ORD can be obtained with
optional upgrade of the Bureau search capacity.
2.3.7.2 Average Response Time
To complement the 90 percentile response time metrics, average response times will
also be measured and scored starting at FOC. Figure 2.3-19 provides the initial
average response time target values for FOC and FOC+3 years.
Average Target
Response Time
Database
Search Type
Category
Priority
(Minutes)
Penetration
FOC
FOC+3
Print-to-Print
100%
Central
Urgent
1.5
1.2
Central
Normal
2.5
2.0
ORD
Normal
2.5
2.0
Print-to-Mark
100%
Central
Urgent
20
15
Central
Normal
30
22
ORD
Normal
7.5
6
Finger Mark-to-Print
100% for 15% of
Central
Urgent
3
2
searches and 30%
for 85% of searches
Central
Normal
5.5
4.5
ORD
Normal
4
3
Palm Mark-to-Print
100% for 15% of
Central
Urgent
8
6
searches and 30%
for 85% of searches
# 30%
Central
Normal
10
8
ORD
Normal
5
4
> 30% and # 100%
Central
Normal
30
24
for up to 15% of
Schedule F, Service Level Requirements
57
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
searches
ORD
Normal
10
8
Mark-to-Mark
100%
Central
Urgent
4
3
Central
Normal
6
4
ORD
Normal
4
3
Figure 2.3-19 Average Target Response Times at FOC and FOC+3
In Figures 2.3-19 and 2.3-18, database penetration refers to geographic penetration
(100% penetration for National searches). The response times reflect search
parameters such as finger allowance and priority currently used in operational
environments. Bureau search loads for the ORD are not specified in the DOR.
The metric for average response time will be the percentage deviation from the target
value ie ([average response time target value - measured average response
time]/average response time target value). For each search type, this metric will be
computed for each priority and search category within the search type. The values
will then be averaged to obtain a composite metric value for the search type. The
calculation of the average response times will exclude searches that are directly
effected by the failure of critical components where the Service Failure is being fully
measured under the Service Availability category. Figure 2.3-20 provides the table
for translating the metric values into scores.
Schedule F, Service Level Requirements
58
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Average Response
Average Response
Average Response
Score
Score
Score
Time Metric (%)
Time Metric (%)
Time Metric (%)
+50.0
120
-10.0
96
-70.0
72
+47.5
119
-12.5
95
-72.5
71
+45.0
118
-15.0
94
-75.0
70
+42.5
117
-17.5
93
-77.5
69
+40.0
116
-20.0
92
-80.0
68
+37.5
115
-22.5
91
-82.5
67
+35.0
114
-25.0
90
-85.0
66
+32.5
113
-27.5
89
-87.5
65
+30.0
112
-30.0
88
-90.0
64
+27.5
111
-32.5
87
-92.5
63
+25.0
110
-35.0
86
-95.0
62
+22.5
109
-37.5
85
-97.5
61
+20.0
108
-40.0
84
-100.0
60
+17.5
107
-42.5
83
-102.5
59
+15.0
106
-45.0
82
-105.0
58
+12.5
105
-47.5
81
-107.5
57
+10.0
104
-50.0
80
-110.0
56
+7.5
103
-52.5
79
-112.5
55
+5.0
102
-55.0
78
-115.0
54
+2.5
101
-57.5
77
-117.5
53
0.0
100
-60.0
76
-120.0
52
-2.5
99
-62.5
75
-122.5
51
-5.0
98
-65.0
74
≤ -125.0
50
-7.5
97
-67.5
73
Figure 2.3-20 Average Response Times Score Table
The overall response time service level score for each search type will be computed
Schedule F, Service Level Requirements
59
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
as a weighted average of the scores for the 90 percentile and the average response
times, with a weight of 0.6 for the 90 percentile and a weight of 0.4 for the average
response time.
In addition to the average response time metrics, shadow metrics for 99.9 percentile
response times will be defined and evaluated for potential implementation after FOC.
2.3.7.3 Non-Verified Live ID Response Time
Since Live ID non-verified searches require much faster response than the other print-
to-print searches, separate measurements will be made and scored. Shortly after TOR,
a live ID shadow metric will be defined and evaluated. The initial target value will be
120 seconds. The implementation and scoring of the metric for Live ID non-verified
searches will commence as soon as the IDENT1 web server architecture is deployed.
The proposed weighting for its inclusion is shown in Figure 2.3-21.
Search Response Time Weights
Print-to-Print
Print-to-Mark
Mark-to-Print
Mark-to-Mark
Quoted CRO
Paper
Live ID
0.05
0.05
0.10
0.10
0.02
0.03
Figure 2.3-21 Search Response Time Weights
On implementation and scoring of the Live ID non-verified search Service Level
Metric, the weighting between other Service Level metrics within the Search
Performance Service criteria will be adjusted in accordance with Figure 2.3-21 above.
2.3.7.4 Serious Crimes Cache
At FOC the target response time for searches of the serious crimes cache will be
reduced to a target of 5 minutes. This target is for review purposes and will not be
included in the scoring mechanism.
2.3.7.5 Operator Task Times
While end-to-end search response time will continue to be defined from the time a
search is submitted by the Bureau to the time the search results are available for
comparison at the Bureau, significantly improved target operator times for various
Bureau tasks have been defined for audit purposes to ensure improved system
responsiveness. These are shown in Figure 2.3-22.
Time at
Time at
Tasks
Description
TOR
FOC
(seconds)
(seconds)
Print-to-Print Processing
Capture Images
Average number of seconds required to scan a two-
First one
First one 34
in a Batch
sided Ten Print form through the Improvision
78.2 sec
sec Others
scanner, display it to the Ten Print Technician, and
Others 56.6
18 sec
have the technician move the delineation boxes if
sec
Schedule F, Service Level Requirements
60
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Time at
Time at
Tasks
Description
TOR
FOC
(seconds)
(seconds)
needed, and accept the form into the system
Check
Number of seconds that a Ten Print Technician
First one 31
First one 7
Demographics in spends checking demographics for one form in a
sec Others
sec Others 6
a Batch
batch
24 sec
sec
Check Prints
Time to start the check prints function from request
2.1 sec
2.1 sec
initialisation for
to display of the first images
a batch
Check Prints
Time per suspect finger for an FPO to inspect and
12 sec
12 sec
Visual
decide on action
Sequence/Quality
Check
Check Prints
Time for an FPO to replace a poor quality rolled
6.3 sec
6.3 sec
Replace a roll
impression with a flat per finger replace
with a flat
Check Prints
Time for an FPO to exchange two fingers placed in
6.3 sec
6.3 sec
Exchange
the wrong box
Fingers
Check Prints
Time for an FPO to manually edit a finger (per
80 sec
80 sec
Fingerprint Edit
finger edited)
Check Prints
Time for an FPO to interchange hands that have been 50 sec
10 sec
Exchange Hands place in the wrong position
Check Prints
Time to accept a form and if any remain in the batch, 0.8 sec
0.8 sec
Accept and Next display the next form
Set parameters
This function is automated
0 sec
0 sec
for default search
Request Non-
Time for Ten Print Technician to set parameters and 50 sec
30 sec
Default Search
start non-default search
Print-to-Print
Time for a Ten Print Technician or FPO to view a
First one
First one
Respondent
respondent image against an enquiry image during a 29.5 sec
29.5 sec
image viewing
first level check (Compare 1) when viewed in a
Others 26
Others 26
time (first check) batch
sec
sec
Print-to-Print
Time for a FPO to view a respondent image against
First one
First one
Respondent
an enquiry image during a first level check
64.4 sec
64.4 sec
image viewing
(Compare 1) when viewed in a batch
Others 60.9
Others 60.9
time (second
sec
sec
check)
Schedule F, Service Level Requirements
61
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Time at
Time at
Tasks
Description
TOR
FOC
(seconds)
(seconds)
Finger Mark-to-Print Processing
Case Control
Time for FPO Create a case and enter details
58.2 sec
58.2 sec
Capture Lift
Average time for FPO to capture a lift image
29 sec
11.5 sec
Clip Mark from
Time for FPO to define and clip the image for one
48 sec
10 sec
Lift
mark from a lift
Edit Image
Average Time for FPO to manually edit minutiae for 155 sec
155 sec
an image
Request Search
Time for FPO to set parameters and start search
14.6 sec
5 sec
Mark-to-Print
Average Time for FPO to view a single finger ten-
First one
First one
Respondent
print respondent for a mark to print search at the first 27.2 sec
18.2 sec
Viewing Time
level
Others 10.2
Others 8.2
(Compare 1)
sec
sec
Mark-to-Print
Average Time for FPO to view a single finger ten-
First one
First one
Respondent
print respondent for a mark to print search at the
98.8 sec
78.5 sec
Viewing Time
second or third level
Others 85.8
Others 71.5
(Compare 2/3)
sec
sec
Palm Mark-to-Print Processing
Case Control
Time for FPO Create a case and enter details
N/A
58 sec
Capture Lift
Average time for FPO to capture a lift image
N/A
11.5 sec
Capture Mark
Time for FPO to define and clip the image for one
N/A
20
from Lift
mark from a lift
Edit Image
Average Time for FPO to manually edit minutiae for N/A
280 sec
an image
Request Search
Time for FPO to set parameters and start search
N/A
5 sec
Mark-to-Print
Average Time for FPO to view a single finger ten-
First one
Others 30.2
Respondent
print respondent for a mark to print search at the first 40.2 sec
sec
Viewing Time
level
(Compare 1)
Mark-to-Print
Average Time for FPO to view a single finger ten-
First one 78
Others 81.2
Respondent
print respondent for a mark to print search at the
sec Others
sec
Viewing Time
second or third level
91.2 sec
(Compare 2/3)
Figure 2.3-22 Operator Task Times
Schedule F, Service Level Requirements
62
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Note that the “TOR times” are applicable after the Linux workstations have been
deployed. This is scheduled to occur shortly after TOR.
By FOC, Figure 2.3-22 will be updated to include operator task times for Print to
Mark searches, searches against the Police Elimination Database and the Operational
Response Database, and the Livescan processes. In addition a tabulation of search
response times for Police Elimination Database, the Operational Response Database
and Serious Crime Cache searches will be added for audit and reporting purposes as
needed rather than SLR scoring, if agreed between the Contractor and the Authority.
Schedule F, Service Level Requirements
63
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
2.4
Technical Support Service
Technical Support Service will be measured in terms of response time for trouble calls made to the
Contractor Help Desk. Response time is defined as the Contractor’s level of responsiveness to any
Bureaux, Central, DRS and training incident, see Figure 2.4-1 for the definition of each response
level.
Response
Title
Description
Level
1
Receive incident call
Answer the incoming call at the help desk and resolve or
escalate.
2
Respond with Status
Resolved incident over the phone or escalate to next level.
Update
3
Restore service to faulty
Site support sent to correct incident and/or restore service
item
to sufficient IDENT1 functional and performance
capability to process their normal workload within the
measures defined within this SLR.
4
Global Changes
Deploy hardware or software fixes to all sites to restore
IDENT1 operational functionality. This response level is
not applicable to all incidents, but only those incidents
which cause a loss of operational services (as defined in
Section 2.1 of this document) at Central, Bureaux, or
Training Centres.
Figure 2.4-1 Definition of Response Levels
The Help Desk, at the first level, will determine the severity of the incident through the information
provided by the caller and in accordance with the Help Desk calling procedures defined in the
IDENT1 Maintenance Plan. In some cases, where sites have only one workstation, it may be
determined that those sites may not employ a level 1 fault severity but fall into a level 2 fault severity.
Figure 2.4-2 defines each fault severity rating for IDENT1.
Fault
Severity
Title
Description
Level
0
Non-fault
All system or service matters that do not equate to a fault.
This will include all calls to the Help Desk that do not fall
into one of the other severity levels.
1
Single User Fault or Non
Faults affecting:
Critical Fault
•
a single user account or workstation
•
non-critical equipment as defined in Section 6.1 of
this SLR
•
non-essential IDENT1 software capabilities that
do not cause a loss of Site Services.
2
Bureau Site Fault
Faults causing a loss of Bureaux Site services involving
critical IDENT1 equipment as defined in Section 6.1 or
Schedule F, Service Level Requirements
64
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
essential IDENT1 software capabilities.
3a 3b
(a) Central Segment -
Faults causing a loss of Central Segment - Primary
Primary Fault (b) Central
services involving critical IDENT1 equipment as defined
Segment - DRS Fault
in Section 2.1 or essential software capabilities Faults
causing a loss of DRS services involving critical IDENT1
equipment as defined in Section 2.1 or essential IDENT1
software capabilities
4
Global Fault
Faults causing a loss of services between Central and
Bureaux Sites, or between Central and PNC, due to one or
more faults.
Figure 2.4-2 Definition of Fault Severity Levels
Non-fault Calls
Some calls received by the Help Desk may not relate to an actual fault but refer to other system or
service matters. Examples are queries about use of the system, user account changes and calls in
connection with user training. The Help Desk will record these calls under one of the following
categories - for analysis and reporting purposes:
• Unconfirmed faults (i.e. problems/suspected faults which, upon examination, do not represent a
failure of the system)
• User queries
• User training (calls in connection with training centre or other recognised training sessions).
Based on the Response Levels defined in Figure 2.4-1 and the Fault Severity Levels defined in Figure
2.4-2, the target response times in Figure 2.4.3 have been established.
Fault Severity Levels
1
2
3
4
0 Non Fault
1 Minute
30 Minutes
Not Applicable
Not
Applicable
1 Single User Fault/Non Critical 1 Minute
30 Minutes
24 Hours
6 Months
Fault
2 Bureau Site Fault
1 Minute
30 Minutes
4 Hours
6 Months
3a Central Segment - Primary
1 Minute 1
30 Minutes
2 Hours 8 Hours
6 Months 6
Fault 3b Central Segment - DRS Minute
30 Minutes
Months
Fault
4 Global Fault
1 Minute
30 Minutes
3 Hours
6 Months
Figure 2.4-3 Target Response Times
Restoration of Service is defined as restoration of sufficient IDENT1 functional capability to enable
Bureaux to process their normal workload within the measures defined within this SLR. This will be
agreed between the Help Desk and the bureau. Technical Support Service will be measured with
respect to response times for each Response Level to trouble calls as defined in Figure 2.4-1. In view
Schedule F, Service Level Requirements
65
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
of the variable nature of Non-faults (Fault Severity Level 0), Target Response Times for Response
Levels 3 and 4 will not be applicable and will not be scored
Once the response times for each level is computed, a SCORE can be calculated for each response
level. This score is calculated on a monthly basis and is illustrated in Section 2.4.1 through 2.4.4. To
compute an OVERALL SCORE for Technical Support Service each response level measure Score
will be computed and applied to Figure 2.4-4. The following steps are used to compute an Overall
Service Support Score:
1. The SCORE for each Measure within Technical Support Service is applied to Figure 2.4-4. These
scores are computed from the scoring tables defined in Section 2.4.1 through 2.4.4 of this SLR.
2. The Sum of the Scores is calculated and then divided by the total number of Measure Scores to
obtain an Overall Service Support Score as shown in Figure 2.4-4.
Measure
SCORE
OVERALL SCORE
Receive Incident Call
Respond with Status Update
Restore Service to Faulty Item
Global Changes
Overall Service Support Score
Figure 2.4-4 Technical Support Service Score
The OVERALL Service Support SCORE will be applied to Figure 4.1-1 for SLR reporting.
2.4.1
Receive Incident Call
The first level of maintenance response is provided by the Contractor’s Help Desk
which is located at the Central Site. For any Bureau question or maintenance
problem that is beyond user’s level of expertise, an incident call, which can originate
as a telephone call, electronic mail, or FAX, will be placed to the Contractor’s Help
Desk. Each IDENT1 Site is provided a Problem Reporting Checklist developed for
IDENT1 operations which is defined in the IDENT1 Maintenance Plan. This
Problem Report Checklist includes the required information needed by the Help Desk
to assess the Incident’s severity.
To compute a measure for Incident Calls the following equation and steps will be
used:
Number of Calls Re sponded to in1min
Equation 2.4-1:
Re ceive Incident Call =
×100
Number of Incident calls received
where
(a)
The Number of Incident Calls responded to in 1 minute is defined as the Help
Desk’s answering any incoming incident calls within 1 minute.
Schedule F, Service Level Requirements
66
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
(b)
The denominator of the equation is the total Number of Incident Calls
received at the Help Desk.
The Receive Incident call measure value will be computed using Equation 2.4-1.
Once the value has been computed, a Service Level Score will be selected from the
Scoring Table presented as Figure 2.4-5. In the event that no applicable incident calls
are made to the help desk in the reporting period, a fixed Service Level Score of 100
shall be applied.
Receive
Receive
Receive
Receive
Receive
Incident SCORE Incident SCORE Incident SCORE Incident SCORE Incident SCORE
Call
Call
Call
Call
Call
100
110
89
94
78
83
67
72
56
61
99
108
88
93
77
82
66
71
≤55
60
98
106
87
92
76
81
65
70
97
104
86
91
75
80
64
69
96
102
85
90
74
79
63
68
95
100
84
89
73
78
62
67
94
99
83
88
72
77
61
66
93
98
82
87
71
76
60
65
92
97
81
86
70
75
59
64
91
96
80
85
69
74
58
63
90
95
79
84
68
73
57
62
Figure 2.4-5 Receive Incident Call Scoring Table
Once a Service Level Score has been selected, it will be applied to the Technical
Support Service table, Figure 2.4-4, in order to compute an overall support service
score.
2.4.2
Respond With Status Update
Once the Incident call details are entered into the Help Desk Management System, the
Help Desk Operator will then try to either resolve the incident call or escalate the
incident as follows:
(a)
Provide telephone support using a series of scripts residing within the Help
Desk Management System. If telephone support provides a solution for the
Incident, and the user concurs with the corrective action, the status
information will be logged within the Help Desk Management System, and
the incident report closed.
Schedule F, Service Level Requirements
67
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
(b)
If telephone support does not provide a satisfactorily resolution, the Help
Desk Operator will then prioritise and categorise the incident, and pass the
incident on to the most appropriate third party provider in accordance with
the IDENT1 Maintenance Plan. The Help Desk Operator will then provide a
status update to the user as to what the next step is in correcting the incident.
(c)
Non-fault calls (Fault Severity Level 0). If the Help Desk Operator is unable
to resolve a Non-fault call over the telephone, the call will be escalated within
the Contractor or Authority service organisation, as appropriate, to be dealt
with - and a status update provided to the user. This will represent close-out
of the call as far as the SLR is concerned. It will be for the person to whom
the call is escalated to respond to the caller and/or agree a suitable response
time - although the Help Desk will continue to track the call until resolved
In computing a value for the “Respond With Status Update” measure the following
equation and steps will be used:
(
{ Number resolved × )
1
.
1
+ Number statused}
Equation 2.4-2:
Re spond with StatusUpdate =
×100
Number of Incident calls
where
(d)
The Number resolved and the Number statused is defined as the Help Desk
either resolving the incident or escalating the incident and providing the
Bureau updated status within 30 minutes. Calls that are resolved within 30
minutes are awarded a premium by multiplying the Number resolved by 1.1
(e)
The denominator of the equation is the total Number of Incident Calls
received at the Help Desk.
The Respond with status update measure will be computed using Equation 2.4-2.
Once a value has been computed, a Service Level Score will be selected from the
Scoring Table presented as Figure 2.4-6. In the event that no applicable status updates
are required by the help desk in the reporting period, a fixed Service Level Score of
100 shall be applied.
Respond
Respond
Respond
Respond
Respond
Respond
with
with
with
with
with
with
SCORE
SCORE
SCORE
SCORE
SCORE
SCORE
Status
Status
Status
Status
Status
Status
Update
Update
Update
Update
Update
Update
110
120
99
104
88
93
77
82
66
71
55
60
109
118
98
103
87
92
76
81
65
70
108
116
97
102
86
91
75
80
64
69
107
114
96
101
85
90
74
79
63
68
106
112
95
100
84
89
73
78
62
67
105
110
94
99
83
88
72
77
61
66
104
109
93
98
82
87
71
76
60
65
Schedule F, Service Level Requirements
68
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
103
108
92
97
81
86
70
75
59
64
102
107
91
96
80
85
69
74
58
63
101
106
90
95
79
84
68
73
57
62
100
105
89
94
78
83
67
72
56
61
Figure 2.4-6 Respond with Status Update Scoring Table
Once a Service Level Score has been selected, it will be applied to the Technical
Support Service table, Figure 2.4-4, in order to compute an overall support service
score.
2.4.3
Restore Service to Faulty Item
For those Incident calls not resolved by the Help Desk, escalated to a third party
provider, and a status update provided to the user, as discussed in Section 2.4.2, a
third party technician will be dispatched to the site reporting the incident. The
technician will follow the response procedures defined in the IDENT1 Maintenance
Plan. To restore service of a hardware or software faulty item, the dispatched
technician will perform the following maintenance tasks:
(a)
Verification
(b)
Fault Isolation
(c)
Remove/Replace or Repair the faulty item (hardware or software) as
appropriate
(d)
Operational Check-out.
Once the hardware or software fault has been corrected, the on-site technician will co-
ordinate the incident call close-out with the Site personnel and the Contractor’s Help
Desk, prior to leaving the site.
Restore Service to a Faulty Item measure may be one of five (5) severity levels as
defined in Figure 2.4-2. The response time for each is slightly different, leading to
slightly different values. In computing a value for each severity level the following
equation and steps will be used five (5) times, once each for each severity level:
Number of Incident calls resolved
Equation 2.4-3:
Re store Serviceto a Faulty Item =
×100
Number of Incident calls escalated
where
(e)
The Number of Incident Calls resolved is defined in terms of those calls
resolved within the times shown in Figure 2.4-7. The time taken to resolve a
call will be measured from the moment the third party provider receives a call
from the Contractor’s Help Desk until the moment when all parties involved
agree that the incident call has been resolved and the incident is closed. The
threshold for the numerator is based upon the severity of the faulty item as
defined in Figure 2.4-2. In the event of a disaster at the Primary Site and the
use of the DRS to support IDENT1 operations, the 2 hour restore service to
faulty item value shall apply to the DRS.
Schedule F, Service Level Requirements
69
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Fault Severity
Restore service to faulty item
1 Single User Fault/Non Critical Fault
24 Hours
2 Bureau Site Fault
4 Hours
3a Central Segment Primary Fault
2 Hours
3b Central Segment - DRS
8 Hours
4 Global Fault
3 Hours
Figure 2.4-7 Restore Service to Faulty Item Response times
(f)
The denominator of the equation is the total Number of Incident Calls
escalated by the Help Desk for each of the severity levels defined in Figure
2.4-2.
The Restore Service to a Faulty Item measure will be computed using Equation 2.4-1
five (5) times, once for each severity level defined in Figure 2.4-2. The value
calculated for each severity level will be applied to Figure 2.4-8.
Schedule F, Service Level Requirements
70
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Severity Level
% Single
Weighted
User
%
Score
Average
Fault/Non
% Bureau
Global
% Central Segment Fault
Critical
Site Fault
Fault
Fault
Primary
DRS
Weight
.25
.25
.20
.05
.25
Restore
Service to
Faulty Item
Figure 2.4-8. Measured Restore Service to Faulty Item Score
The five (5) values will be multiplied by their weights and summed to produce a
weighted average for the Restore Service to Faulty Item measure.
Once an Average has been computed, a Service Level Score will be selected from the
Scoring Table presented as Figure 2.4-9 and applied to Figure 2.4-8. This Service
Level Score will be carried forward and applied to Figure 2.4-4. In the event that no
applicable service incidents are made during the reporting period, a fixed Service
Level Score of 100 shall be applied.
Restore
Restore
Restore
Restore
Restore
Service to SCORE Service to SCORE Service to SCORE Service to SCORE Service to SCORE
Faulty Item
Faulty Item
Faulty Item
Faulty Item
Faulty Item
100
110
89
94
78
83
67
72
56
61
99
108
88
93
77
82
66
71
≤55
60
98
106
87
92
76
81
65
70
97
104
86
91
75
80
64
69
96
102
85
90
74
79
63
68
95
100
84
89
73
78
62
67
94
99
83
88
72
77
61
66
93
98
82
87
71
76
60
65
92
97
81
86
70
75
59
64
91
96
80
85
69
74
58
63
90
95
79
84
68
73
57
62
Figure 2.4-9 Measured Restore Service to Faulty Item Scoring Table
In the event of a disaster at the primary site and the use of the DRS to support
Schedule F, Service Level Requirements
71
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
IDENT1 operations, no % Central Segment Fault-Primary will be computed. Rather
the % Central Segment Fault of the DRS site will be used with a weight of .25 in
Figure 2.4-8. If there is a disaster at the DRS, the % Central Segment Fault- Primary
value will be computed as normal and a weight of .25 will be used in Figure 2.4-8.
2.4.4
Global Changes
In addition to incident calls related to problems discussed in Section 2.4.3, some
incidents may be referred to the Contractor’s engineers for resolution in future system
releases. These escalated incidents will be considered global changes as they
constitute changes that need to be applied to IDENT1 as a whole. Incidents that affect
operational services at Central, Bureaux, or Training Centres will be resolved by the
Contractor’s development/test group and included in new releases within the time
scales specified in Figure 2.4-3. Incidents that do not affect operational services will
not be included in global changes and will not subject to the time scales specified in
Figure 2.4-3. The status of all applicable escalated incidents will be tracked
separately, with status reporting on a monthly basis, as a minimum, and a resolution
target as defined in Figure 2.4-3.
To compute a measure for Global Changes the following equation and steps will be
used:
Number of Incident calls resolved
Equation 2.4-4:
Global Changes =
×100
Number of Incident calls
where
(a)
The Number of Incident Calls resolved is defined as the number of calls
escalated to the Contractor’s development/test group and resolved, installed,
checked out, and closed out in 6 months.
(b)
The denominator of the equation is the total Number of Incident Calls
escalated to the Contractor’s development/test group for resolution.
The Global Changes measure will be computed using Equation 2.4-4.
Once a value has been computed, a Service Level Score will be selected from the
Scoring Table presented as Figure 2.4-10. The resulting score will be applied to
Figure 2.4-4. In the event that no applicable incidents are outstanding or scheduled to
be fixed at the time of the Service Review, a fixed Service Level Score of 100 shall
be applied to Global Changes.
Global
Global
Global
Global
Global
SCORE
SCORE
SCORE
SCORE
SCORE
Changes
Changes
Changes
Changes
Changes
100
110
89
94
78
83
67
72
56
61
99
108
88
93
77
82
66
71
≤55
60
98
106
87
92
76
81
65
70
97
104
86
91
75
80
64
69
Schedule F, Service Level Requirements
72
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
96
102
85
90
74
79
63
68
95
100
84
89
73
78
62
67
94
99
83
88
72
77
61
66
93
98
82
87
71
76
60
65
92
97
81
86
70
75
59
64
91
96
80
85
69
74
58
63
90
95
79
84
68
73
57
62
Figure 2.4-10 Measured Global Changes Scoring Table
2.4.5
Future Changes for Service Management
No changes of the target values for the helpdesk response times are currently planned.
The mechanism for faster response time will continue to be the incentive built into the
SLR scoring structure.
The focus for this service level criterion after TOR will be on the definition,
evaluation of shadow metrics that reflect user service satisfaction both from a
helpdesk perspective and the overall service support. Depending on the evaluation
results, it is a goal that at or prior to FOC, one or more user oriented metrics will be
implemented and integrated into the IDENT1 SLR structure described in Section 1.0
2.5
Search Accuracy
AFR Accuracy will be measured in terms of matcher performance. Matcher performance is
defined as the systems’ accuracy when performing a search. The Contractor and the
Authority will establish a background database called a Quality Assurance (QA) Collection
database located at the Central Site. The objective of this QA Collection database is to ensure
that search accuracy is not degraded due to changes, enhancements, or upgrades to system
equipment or matcher algorithms. This database will consist of Ten Prints and Marks,
enquiry Ten Prints and Marks, identification of the correct match (ident) for each enquiry, and
sets of respondent IDs and scores from previous runs. Maintaining the ident information
allows the QA run results to be measured in terms of the position on each respondent list.
Improvements in accuracy from one QA run to the next is then measured as the amount that
the ident has moved towards the top of the respondent list.
In measuring matcher performance, a limited number of known searches will be run. These
known searches are run to ensure that the system provides the same score as established with
the baseline database, should no changes have been made. In addition, if changes have been
made the QA run of known searches should yield the same results and not degrade the
baseline search accuracy. See Section 2.5.1 of this SLR for a description of how AFR
Accuracy will be measured.
2.5.1
Matcher Performance
Matcher Performance Accuracy will be assessed monthly from each QA Collection
database run. A QA run consists of running searches using each enquiry item. The
top portion of the respondent list is retained as the baseline, see Figure 2.5-1. The
Schedule F, Service Level Requirements
73
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
initial baseline QA Collection database will be implemented from data obtained at
previous AFR trials and augmented with data to be provided by the Authority.
Background
Respondent IDs &
Search Type
Enquires
Database
Scores to Save
Print-to-Print
500
50,000
Top 3 (All 3 fingers
searched)
Mark-to-Print
400
50,000
Top 15
Print-to-Mark
125
5,000
Top 10
Mark-to-Mark
125
5,000
Top 5
Figure 2.5-1 Search Type Data Saved
Search accuracy will be measured for each search type as defined in Figure 2.5-1.
Data obtained for each search type will be compared for both the QA run (Test Set)
and the existing baseline (Baseline Set) as shown in Figure 2.5-2.
The data resulting from each search type will be a percentage value of idents in the
Nth position as shown in Figure 2.5-2, based on the definitions of each search type
below:
1. Print-to-Print: N is 1, ≤ 2, ≤ 3, >3
2. Mark-to-Print: N is 1, ≤ 10, ≤ 15, >15
3. Print-to-Mark: N is 1, ≤ 10, >10
4. Mark-to-Mark: N is 1, ≤ 5, >5.
Schedule F, Service Level Requirements
74
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Matcher Quality Assurance Report
Test Data Set:ref_num, run_num, date
Baseline Data Set: ref_num, run_num, date
Print to Print
Test Set
Baseline Set
Idents in top position
00%
00%
Idents in top 2 positions 00%
00%
Idents in top 3 positions 00%
00%
Idents not in top 3 positions
00% 00%
Mark to Print
Test Set
Baseline Set
Idents in top position
00%
00%
Idents in top 10 positions 00%
00%
Idents in top 15 positions 00%
00%
Idents not in top 15 positions
00% 00%
Print to Mark
Test Set
Baseline Set
Idents in top position
00%
00%
Idents in top 10 positions 00%
00%
Idents not in top 10 positions
00% 00%
Print to Mark Serious Crime Cache
Test Set
Baseline Set
Idents in top position
00%
00%
Idents in top 10 positions 00%
00%
Schedule F, Service Level Requirements
75
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Idents not in top 10 positions
00% 00%
Mark to Mark
Test Set
Baseline Set
Idents in top position
00%
00%
Idents in top 5 positions 00%
00%
Idents not in top 5 positions
00% 00%
Figure 2.5-2 Matcher QA Report
The monthly assessment will be run based on one of the following:
(a)
Should there be no changes to the baseline IDENT1 design configuration, a
subset of the enquiries will be searched for each search type to ensure that
percentages for both the Baseline Set and Test Set remain the same.
(b)
For Intermediate cases where a new version of the UNIX operating system,
new version of the compiler is installed, there may be slight changes (i.e.,
rounding off changes). These will be considered as scores remaining the
same.
(c)
For situations where a known IDENT1 configuration is brought down and
then rebuilt, a subset of the enquiries will be searched for each search type to
verify that percentages for both the Baseline Set and the Test Set remain the
same.
(d)
For the situation where a matcher upgrade is installed, the entire set of
searches would be re-searched. For the upgrade to pass, accuracy must not
be degraded for any search type. For Print-to-Print, accuracy must not
decrease for top 1, top 2 and top 3 respondent positions. For Mark-to-Print,
accuracy must not decrease for top 1, top 10, and top 15 positions. For Print-
to-Mark, accuracy must not decrease for top 1 and top 10 positions. For Print-
to-Mark Serious Crime Cache, accuracy must not decrease for top 1 and top
10 positions. For Mark-to-Mark, accuracy must not decrease for top 1 and
top 5 positions.
These criteria may be relaxed with Authority approval. If the QA run passes, then the
new set of scores becomes the new Score baseline.
2.5.2
Matcher Score
In computing a monthly measure for AFR Performance the data collected and
compared in the Matcher QA Report, shown in Figure 2.5-2 within this SLR, will be
used. The Service Level Score will be determined by comparing the Baseline Set
percentages and the Test Set percentages for each measure in Figure 2.5-3:
Schedule F, Service Level Requirements
76
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Test Set
Baseline Set
Difference
Score
Print to Print
•
Idents in top position
0%
0%
0.0%
•
Idents in top 2 positions
0%
0%
0.0%
•
Idents in top 3 positions
0%
0%
0.0%
•
Idents not in top 3 positions
0%
0%
0.0%
Mark to Print
•
Idents in top position
0%
0%
0.0%
•
Idents in top 10 positions
0%
0%
0.0%
•
Idents in top 15 positions
0%
0%
0.0%
•
Idents not in top 15
0%
0%
0.0%
positions
Print to Mark
•
Idents in top position
0%
0%
0.0%
•
Idents in top 10 positions
0%
0%
0.0%
•
Idents not in top 10
0%
0%
0.0%
positions
Print to Mark Serious Crime
Cache
•
Idents in top position
0%
0%
0.0%
•
Idents in top 10 positions
0%
0%
0.0%
•
Idents not in top 10
0%
0%
0.0%
positions
Mark to Mark
•
Idents in top position
0%
0%
0.0%
•
Idents in top 5 positions
0%
0%
0.0%
•
Idents not in top 5 positions
0%
0%
0.0%
Figure 2.5-3 Matcher QA Report
The Difference will be computed by subtracting the Baseline Set percentage from the
Schedule F, Service Level Requirements
77
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Test Set percentage. If the Difference is 0.0 then a score of 100 will be applied to
Figure 2.5-3 and carried forward to Figure 4.1-1 for each Bureau. If a Difference
other than 0.0 is obtained for any of the measures in Figure 2.5-3, then a score of 0
will be applied to Figure 2.5-3 and carried forward to Figure 4.1-1 for each Bureau.
2.5.3
Future Changes for Search Accuracy
Significant changes for the search accuracy service level criterion will be made after
TOR. These include the benchmark measurement of accuracy improvement and the
integration of the results with the existing quality control metric to determine the
service level score for the search accuracy criterion.
2.5.3.1 Accuracy Improvement Metrics
To support the need for continuing accuracy improvement for IDENT1, search
accuracy metrics are defined to measure planned improvements using benchmarking
every 2 or 3 years. To facilitate the benchmarking process, three search types, print-
to-print, finger mark-to-print, and palm mark-to-print are used to represent the overall
search accuracy of the system. In all cases, the most important Top 1 accuracy will be
used as the metric. Figure 2.5-4 shows the planned evolution of the target accuracy
levels from FOC to FOC+1 and FOC + 4 years.
Search Type
Metrics
Target Accuracy Levels
FOC
FOC+1
FOC+4
Print-to-Print
Top 1
T1=99.8%
No Change
No Change
Finger Mark-to-
Top 1
T2=54.2%
T2+5%
T2+10%
Print
Palm Mark-to-
Top 1
T3=76.2 %
T4+3%
T4+5%
Print
Figure 2.5-4 Search Accuracy Target Values
Where the target accuracy levels for the searches are specified in terms of the
IDENT1 benchmark results, T1, T2, T3 and T4 as defined below:
T1 = Northrop Grumman IDENT1 ITT Benchmark Print-to-Print accuracy
T2 = Northrop Grumman IDENT1 ITT Benchmark Finger Mark-to-Print
accuracy
T3 = Northrop Grumman IDENT1 ITT Benchmark Palm Mark-to-Print
accuracy based on a 80K subject dataset
T4 = Northrop Grumman IDENT1 Benchmark Palm Mark-to-Print accuracy on
a 1 million subject dataset measured at FOC
Note that the target accuracy levels assume searches against a 1 million-person
background database unless indicated otherwise. For the Mark to Print (Palm) search
accuracy measurement there does not currently exist a background data set of 1
million subjects for benchmarking purposes. The Accuracy figure for T4 for a 1
million database at FOC will be measured concurrent with the search accuracy
Schedule F, Service Level Requirements
78
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
measurement on the 80k subject database at FOC as part of the same benchmark. All
benchmarking for subsequent milestones for the search accuracy measurements based
on T4 will use a 1 million subject database.
For each search type, the search accuracy for SLR purposes is calculated as the
percentage of searches for which the specified mate (or any duplicate) is identified in
the top one position, ie (Number of top one Idents/Total number of searches) * 100.
For finger mark-to-print searches, the accuracy improvement at FOC will be based
primarily on the fusion of SAGEM and NAFIS matcher results. Improvements at
FOC+1 will derive from the use of multi-rolls and flats. Note that although these are
FOC functionality, they will be benchmarked at FOC+1 to ensure that the benchmark
background database will contain adequate representation of these prints. Further
accuracy improvements at FOC+4 will be based on future algorithm enhancements.
The benchmark search accuracy measurement will be made against a standard 1
million subject background database, for example, the IDENT1 benchmark database.
The same edited mark search prints will be used to eliminate variations due to
operators. To avoid algorithm tailoring, additional tests can be conducted using
marks that are identified during operation with the background database appropriately
seeded with the identified mates.
The approach to each search accuracy benchmarking event will be agreed between
the Contractor and Authority and documented prior to the start of the associated
benchmark.
Further accuracy improvements beyond the target values specified in Figure 2.5-4 for
Mark to Print Searches will be through the scoring tables that provide appropriate
incentives for superior performance. Accuracy improvement for Print – Print Search
Accuracy are not readily distinguishable within the constraints of these benchmarks
and appropriate confidence intervals. The T1 Accuracy will be validated at FOC
through the benchmark test. Successful completion of this benchmark, within
confidence limits agreed as part of the benchmark, will result in the re-base lining of
the QA test as described in section 2.5.1. Benchmarking events for Print to Print
search accuracy at milestones after FOC will be used for information purposes only.
The scoring tables, for mark-to-print searches are provided in Figure 2.5-6. The table
is normalized to provide a score of 100 for the specified target accuracy level for the
specific milestone.
Finger Mark-to-Print
Score
Mark-to-Print Accuracy
Score
Accuracy
TL > 5.0%
120
TL - 2.0%
92
TL + 5.0%
120
TL - 2.5%
90
TL + 4.5%
118
TL - 3.0%
88
TL + 4.0%
116
TL - 3.5%
86
TL + 3.5%
114
TL - 4.0%
84
TL + 3.0%
112
TL - 4.5%
82
Schedule F, Service Level Requirements
79
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
TL + 2.5%
110
TL - 5.0%
80
TL + 2.0%
108
TL - 6.0%
75
TL + 1.5%
106
TL - 7.0%
70
TL + 1.0%
104
TL - 8.0%
65
TL + 0.5%
102
TL - 9.0%
60
Target Level (TL)
100
TL - 10%
55
TL - 0.5%
98
TL - 11%
50
TL - 1.0%
96
< (TL - 11%)
50
TL - 1.5%
94
Figure 2.5-6 Mark-to-Print Accuracy Scoring Tables
Schedule F, Service Level Requirements
80
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
The scoring table for palm mark-to-print search accuracy is provided by Figure 2.5-7
Palm Mark-to-Print
Score
Mark-to-Print Accuracy
Score
Accuracy
TL > 5.0%
120
TL - 2.0%
92
TL + 5.0%
120
TL - 2.5%
90
TL + 4.5%
118
TL - 3.0%
88
TL + 4.0%
116
TL - 3.5%
86
TL + 3.5%
114
TL - 4.0%
84
TL + 3.0%
112
TL - 4.5%
82
TL + 2.5%
110
TL - 5.0%
80
TL + 2.0%
108
TL - 6.0%
75
TL + 1.5%
106
TL - 7.0%
70
TL + 1.0%
104
TL - 8.0%
65
TL + 0.5%
102
TL - 9.0%
60
Target Level (TL)
100
TL - 10%
55
TL - 0.5%
98
TL - 11%
50
TL - 1.0%
96
< (TL - 11%)
50
TL - 1.5%
94
Figure 2.5-7 Palm Accuracy Scoring Table
2.5.3.2 Integration with Quality Control Metric
As previously described, the NAFIS quality control metric on search accuracy will be
retained for IDENT1. This metric measures rank consistency using a small database
to ensure software and system changes will not adversely impact the search
performance. For IDENT1, data on score consistency though not used for scoring
purposes, will also be made available for review purposes.
The integration of the accuracy improvement metrics and the existing QC metric will
be accomplished in the following manner. Currently, the QC metric is measured
monthly. Depending on if the consistency check is successful or not, a target score of
100 or zero is assigned. With the inclusion of the accuracy improvement metrics, this
scheme will be retained except the target score will be determined by the accuracy
levels and therefore scores achieved during the last benchmark for the T2, T3 and T4
accuracy measures, as above in Figure 2.5-4.
For each accuracy benchmark, separate scores will be calculated for finger mark-to-
Schedule F, Service Level Requirements
81
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
print and palm mark-to-print searches based on the measured accuracy levels. For the
print to print search accuracy a score of 100 will be assigned. The overall benchmark
score is then calculated as the weighted sum of these individual scores. The weights
for the print-to-print, finger mark-to-print and palm mark-to-print scores are 0.1, 0.6
and 0.3 respectively (these weight will be re-normalized if one or more of the search
types is not benchmarked). This benchmark score will serve as the target score for
monthly QC until it is modified during the next benchmark.
The monthly derived QC score using this target value will be the service level score
for the search accuracy criterion. A 10% weight will be used for this Service Criteria
score in calculating the overall service level score.
Schedule F, Service Level Requirements
82
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
3
SHADOW METRICS
Shadow metrics are new metrics proposed to enable the SLR to achieve a better reflection of system
performance, business value and user satisfaction. As previously described, they are the means for
evolving the initial IDENT1 system-centric SLA structure to a structure that also measures and
addresses business values and user needs.
Prior to the implementation of these new metrics, they need to be carefully measured and evaluated to
determine the utility and also to establish the baseline values. Figure 3.5-1 provides the initial list of
shadow metrics that have been identified in each metric category and their implementation time
frames.
Category
Shadow Metrics
Implementation Timeframe
System Performance
Live ID Response Time
No later than FOC
Planned Downtime
No later than FOC
Interface Metrics
Post FOC
Business Values
Number of Idents
Post FOC
Number of Idents per Search
Post FOC
or Labor Hour
User Effort per Transaction or
Post FOC
Business Thread
User
Helpdesk Service Satisfaction
No later than FOC
Overall Service Support
No later than FOC
Satisfaction
Figure 3.5-1 Initial Shadow Metrics
The listed shadow metrics will be further defined, agreed with the Authority and evaluated after TOR
for potential implementation depending on the evaluation results. The process for evaluating,
implementing the shadow metrics and changes to the SLR will be in accordance with the change
control process defined in Part 1 of this Schedule.
Schedule F, Service Level Requirements
83
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
4
SERVICE SCORE
4.1
SERVICE PROVISION SCORES
This section defines how the various Service Level Metrics measured for each individual IDENT1 site
or centrally are summated to derive an overall Service Scores for individual sites and how the Overall
Score for the Service is derived from the individual site Service Scores.
The Monthly Service Charge for the IDENT1 Service is detailed in Schedule E (Pricing). The
adjustment to the Monthly Service Charge (Service Credits or Service Incentivisation Payments) are
based on the quality of service delivered to the individual IDENT1 Sites (Police Force and Training
Centres) and measured under the SLR.
Monthly, the Contractor will present the Service Score associated with each Service Level Metric and
the summary Service Score for each IDENT1 site. The computation of the Service Score from the
Measure for each Service Level Metric is detailed in Section 2.0, step #1 and the computation of the
Service Score for the individual IDENT1 site is detailed in Section 2.0, step #2. The calculated
Service Scores will be presented in tabular format based on Figure 4.1-1. The agreed weightings
between each Service Level Metric and Service Criteria are detailed in the header of the row of the
figure. In addition, the Contractor will present the individual Service Measures for each Service Level
Metric in tabular form as defined and detailed in Section 2 and 3 of this Part.
The Service Scores for individual IDENT1 sites are taken through to Schedule E (Pricing) to derive
the Service Credit or Service Incentive Payment for each site, which are then aggregated to form the
Actual Service Charge for the month.
The Actual Service Score for the IDENT1 Service is derived from the individual site service scores.
Each individual site is allocated to a tier based on its associated workload/ throughput. The Overall
Score is calculated from the aggregate of Service Scores for the individual site multiplied by the
weighting associated with the tier it is in. The allocation of bureaux to Tiers and the normalised
weighting is detailed in Section 3 of Schedule E (Pricing).
In addition to the summary scoring tabulation and weighting given in Figure 4.1-1, the Site Scores
and Overall Score for the preceding 6 months as illustrated as a spreadsheet in Figure 4.1-2 will be
presented monthly for the purposes of identifying any trigger conditions described in Section 5
Escalation of Service Related Failure of Part 1 of this Schedule and to identify ongoing trends.
Figures 4.1-1 and 4.1-2 will be updated as additional bureau are deployed under the Contract and in
line with the identified changes in the preceding Sections of this Part.
Schedule F, Service Level Requirements
84
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
MONTHLY
Service Level
Operational
Technical
Throughput
System Performance
AFR
SERVICE
Criteria
Availability
Support
SCORE
Central Central
P-M
M- Quoted
Site Ten Print
Mark
P-P P-M
M-P
Service
AFR
Measure
Ao -
Ao -
SCC
M
CRO
A
Throughput Throughput Perf. Perf.
Perf.
Support Accuracy
Primary DRS
o
Perf.
Perf. Perf.
Weight
0.07
0.03 0.10
0.10
0.10
0.10 0.05 0.0 0.10 0.05 0.05
0.15
0.10
Avon + Somerset
Bedfordshire
British Transport
Police
Cambridgeshire
Cheshire
City of London
Cleveland
Cumbria
Customs and
Excise
Derbyshire
Devon +
Cornwall
Dorset
Durham
Dyfed Powys
Essex
Gloucestershire
Greater
Manchester
Gwent
Hampshire
Hertfordshire
Humberside
Schedule F, Service Level Requirements
85
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
MONTHLY
Service Level
Operational
Technical
Throughput
System Performance
AFR
SERVICE
Criteria
Availability
Support
SCORE
Central Central
P-M
M- Quoted
Site Ten Print
Mark
P-P P-M
M-P
Service
AFR
Measure
Ao -
Ao -
SCC
M
CRO
A
Throughput Throughput Perf. Perf.
Perf.
Support Accuracy
Primary DRS
o
Perf.
Perf. Perf.
Weight
0.07
0.03 0.10
0.10
0.10
0.10 0.05 0.0 0.10 0.05 0.05
0.15
0.10
Kent
Lancashire
Leicestershire
Lincolnshire
Merseyside
Metropolitan
N.I.S./Residual
Bureau
Norfolk
North Wales
North Yorkshire
Northamptonshire
Northumbria
Nottinghamshire
South Wales
South Yorkshire
Staffordshire
Suffolk
Surrey
Sussex
Thames Valley
Warwickshire
West Mercia
West Midlands
Schedule F, Service Level Requirements
86
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
MONTHLY
Service Level
Operational
Technical
Throughput
System Performance
AFR
SERVICE
Criteria
Availability
Support
SCORE
Central Central
P-M
M- Quoted
Site Ten Print
Mark
P-P P-M
M-P
Service
AFR
Measure
Ao -
Ao -
SCC
M
CRO
A
Throughput Throughput Perf. Perf.
Perf.
Support Accuracy
Primary DRS
o
Perf.
Perf. Perf.
Weight
0.07
0.03 0.10
0.10
0.10
0.10 0.05 0.0 0.10 0.05 0.05
0.15
0.10
West Yorkshire
Wiltshire
Ext. Transaction
Bureau
Weight
0.25
0.00 0.40
0.00
0.00
0.00 0.00 0.00 0.00 0.00 0.00
0.30
0.05
Hendon Training
Centre
Durham Training
Centre
Monthly Service
Score
Figure 4.1-1 Service Scores for Individual Bureaux
Schedule F, Service Level Requirements
87
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Service
Service
Service
Service
Service
Current
Score
Score
Score
Score
Score
Month
IDENT1 Site Bureaux
Month
Month
Month
Month
Month
(CM)
CM - 5
CM - 4
CM -3
CM - 2
CM - 1
Service
Score
Avon + Somerset
Bedfordshire
British transport Police
Cambridgeshire
Cheshire
City of London
Cleveland
Cumbria
Customs and Excise
Derbyshire
Devon + Cornwall
Dorset
Durham
Dyfed Powys
Essex
Gloucestershire
Greater Manchester
Gwent
Hampshire
Hertfordshire
Humberside
Kent
Lancashire
Leicestershire
Schedule F, Service Level Requirements
88
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
Lincolnshire
Merseyside
Metropolitan
N.I.S./Residual Bureau
Norfolk
North Wales
North Yorkshire
Northamptonshire
Northumbria
Nottinghamshire
South Wales
South Yorkshire
Staffordshire
Suffolk
Surrey
Sussex
Thames Valley
Warwickshire
West Mercia
West Midlands
West Yorkshire
Wiltshire
External Transaction Bureau
Hendon Training Centre
Durham Training Centre
Overall Service Score
Figure 4.1-2. Service Scores for Individual Bureaux
Schedule F, Service Level Requirements
89
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
4.2
Score Adjustment
At the Monthly Service Review, the Contractor shall present to the Authority any score
adjustments that must be made. The charges awarded to the Contractor as a result of these
scores are established in Schedule E (Pricing). Score adjustments may be made as a result of:
Under Performance. If, in any monthly reporting period, the Service Level Score for any of
the five Service Level Criteria at any site is at the minimum allowable Service Level Score,
then the Contractor will not be awarded a charge above 100% in that month even if Service
Level Scores in other Service Level Criteria at other sites are above 100%. This is to ensure
adequate levels of service in all areas.
Excessive Bureau Workload. If, in any monthly reporting period, the workload generated by a
Fingerprint Bureau is above its Bureau Maxima by volume, priority, or geographic search
scope (identified in Section 5), then the bureau’s performance value will be adjusted as
follows:
(a)
Print-to-Print Searches
(i)
If the monthly Bureau Maximum is exceeded, and the achieved performance
is lower than 90, adjust the performance value to 90.
(ii)
If the excess workload is greater than 20% of the Bureau Maximum adjust
the performance value to 100.
(b)
Print-to-Mark, Mark-to-Print, Mark-to-Mark Searches
(i)
If the monthly Bureau Maximum is exceeded, and the achieved performance
is lower than 90, adjust the performance value to 90. For those bureaux that
may perform 100% National searches, this adjustment is made before adding
the ten print National search compensation value.
(ii)
If the excess workload is greater than 20% of the Bureau Maximum, adjust
the performance value to 100.
(iii)
If National searches are more than 5% for any of the search priorities and the
achieved performance is lower than 90, adjust the performance value to 90.
This does not apply to those Bureaux that may perform up to 100% National
searches.
(iv)
If National searches are greater than 10% of the monthly Bureau Maximum,
and the monthly Bureau Maximum has been exceeded adjust the performance
value to 100. This does not apply to those Bureaux that may perform up to
100% National searches.
(v)
Post TOR, this section will be rewritten as the table at Figure 5.2- is updated.
Excessive National Workload. If, in any monthly reporting period, the workload generated by
the Fingerprint Bureaux collectively is above the National Maxima by priority or geographic
search scope (identified in Section 5, Figure 5.2-), then the Service Level Score will be
adjusted to a level of at least 100% for the Search Performance Service Level Criteria. If the
unadjusted aggregate Service Level Score for the Search Performance Service Level Criteria
is above 100%, that computed, unadjusted score will remain.
Other Exceptions.
Schedule F, Service Level Requirements
90
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
(c)
If, in any reporting period, Bureau(x) working practices or other exceptions
detrimentally affect a Service Measure, the Authority may, at its discretion, adjust the
final Service Score (and thus the Service Charges computed as a result of the Service
Score adjustment) of any Service Level Criteria based on a documented claim for
Charge Adjustment provided by the Contractor.
(d)
If, in any reporting period, occurrence of a disaster renders a Bureau inoperable, then
the Disrupted Bureau’s Monthly Service Score (and thus the Service Charges
computed as a result of the Service Score adjustment) will be set to 100. This score
adjustment for the Disrupted Bureau will continue as long as the Bureau is inoperable
and will continue should the Disrupted Bureau choose to operate in a Reciprocal
Bureau arrangement.
(e)
Reciprocal Bureau Operations
(i)
If, in any reporting period, a Disrupted Bureau operates in a Reciprocal
Bureau arrangement, then the Monthly Service Score (and thus the Service
Charges computed as a result of the Service Score adjustment) for the
Reciprocal Bureau processing transactions from both the Disrupted Bureau
and the Reciprocal Bureau will be set to the maximum score. The Monthly
Service Score of the Disrupted Bureau will be set to 100.
(f)
Emergency Bureau Operations
(i)
If, in any reporting period, a Disrupted Bureau operates at the Emergency
Bureau, then the Disrupted Bureau’s Monthly Service Score (and thus the
Service Charges computed as a result of the Service Score adjustment) will
be set to the maximum score.
It will be the responsibility of the Authority to inform a bureau of any excess workload
submitted by the bureau and to make policy changes to reduce the excess workload, if
necessary. The Authority may elect to initiate a formal change with the Contractor to
accommodate the excess workload. In the event that a loss of Availability in one
measurement period (as defined in Section 2.1) causes a bureau to exceed its maximum
workloads in a following period, the Contractor will not receive any service score above
100% in respect of that period
5
IDENT1 WORKLOAD
5.1
National Workload and Sizing Data
The installed IDENT1 configuration will meet the service levels necessary to support actual
Police Force workloads, up to the annual Contract Maximum. IDENT1 may be expanded to
support workloads in excess of the Contracted Maximum, albeit with additional charges.
The annual Contracted Maximum workload for each Bureau is expressed as the number of
Ten Prints and Marks processed by the system and the number retained in the Unified
Collections. These figures are defined below for 2004 (figures are derived from the IDENT1
Detailed Operational Requirement as stated).
Annually, the figures in 5.1-1 and 5.2-1 will be refreshed using the source figure for the new
year given in the DOR tables stated in Figure 5.1-1.
Certain aspects of the IDENT1 service are determined by sizing data inherent in the nature of
fingerprint processing. The service provided, as specified in this SLR, takes into
Schedule F, Service Level Requirements
91
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
consideration these maximum workloads and sizing data. Engineering margins have been
built into the IDENT1 design to accommodate minor variations of the distribution of the total
search load between the search types from that specified in DOR Table 2.18.
Year of Operation
2004
Derivation (ITT DOR references)
Total Ten Print Database Size
7
,529,000 Table 2.7
Total Ten Prints Received and
Processed
2
,880,000 Table 2.13 distributed according to Table 2.14
Total Mark Database Size
1
,238,400 Table 2.10
Total Marks submitted
8
89,500 Table 2.16 distributed according to Table 2.11
Total Searches
7
,570,680 Table 2.17
Unverified Identity Check 4
69,382 Table 2.18
Other P-P 1
,559,560 Table 2.18
Total P-M 3
,278,104 Table 2.18
Total M-P 2
,248,492 Table 2.18
Total M-M 1
5,141 Table 2.18
Figure 5.1-1. National Workload
The maximum National workloads in Figure 5.1-1 are represented as yearly totals for 2004.
The Service Level Criteria presented in Section 2 are based on monthly workload values. The
monthly workload values presented in Section 5.2 of this SLR are assumed to be one twelfth
(1/12) of those shown in Figure 5.1-1. Actual Bureau workload may vary.
Note that the figures in this section provide capacity for Scottish Bureaux at TOR, albeit that
the Bureaux won’t be added to the SLR until a later milestone.
5.2
Bureau Workload
This section defines Bureau workload as a proportion of the National Ten Print and Mark
workloads presented in Section 5.1. Figure 5.2-1 presents the Maximum Monthly Workload
figures for each Bureau. Taken together, these values represent one-twelfth (1/12) of the
annual National Workload as presented in Figure 5.1-1.
Schedule F, Service Level Requirements
92
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
M o n th ly n o n-
T otal P rin ts
v e rified
M on thly P -
M on thly P - T o ta l M arks
M o n th ly M -
M o nth lyM -
S u b m itted
id en tity
P S ea rc h es M
S u b m itte d
P S e arc he s M
M on th ly
ch e ck
(P P m ax )
S e arc he s
M o nthly
(M P m ax )
S ea rc h es
A von & S om erset
5,52 0
1,017
3,379
7,103
4 ,67 0
4,872
33
B edfordshire
6,24 0
1,134
3,769
7,922
81 5
5,434
37
C am bridge sh ire
2,16 0
469
1,560
3,278
1 ,11 2
2,248
15
C he sh ire
3,36 0
313
1,040
2,185
66 7
1,499
10
C ity of Lon don
4 80
39
130
273
44 5
187
1
C le veland
2,88 0
274
910
1,912
1 ,18 6
1,312
9
C um bria
2,40 0
508
1,690
3,551
14 8
2,436
16
D erb ysh ire
3,36 0
391
1,300
2,732
59 3
1,874
13
D evo n a nd C ornw a ll
5,52 0
665
2,209
4,644
81 5
3,185
21
D orset
1,92 0
587
1,949
4,098
66 7
2,811
19
D urh am
2,88 0
274
910
1,912
1 ,18 6
1,312
9
D yfed P ow ys
3,12 0
430
1,430
3,005
22 2
2,061
14
E ssex
4,56 0
548
1,819
3,824
1 ,70 5
2,623
18
G loucestershire
1,92 0
469
1,560
3,278
37 1
2,248
15
G re ate r M an ch ester
1 2,48 0
2,347
7,798
16,391
6 ,07 8
11,242
76
G w ent
2,88 0
430
1,430
3,005
51 9
2,061
14
H am pshire
8,40 0
978
3,249
6,829
1 ,33 4
4,684
32
H ertfordshire
2,88 0
430
1,430
3,005
1 ,92 7
2,061
14
H um be rside
4,08 0
508
1,690
3,551
7 4
2,436
16
K ent
6,96 0
1,056
3,509
7,376
2 ,44 6
5,059
34
Lan ca sh ire
5,76 0
1,330
4,419
9,288
2 ,29 8
6,371
43
Leice stershire
2,64 0
548
1,819
3,824
1 ,33 4
2,623
18
Lin co lnshire
2,16 0
235
780
1,639
22 2
1,124
8
M erseyside
6,72 0
1,291
4,289
9,015
1 ,48 3
6,183
42
M etropolita n
2 2,32 0
6,689
22,224
46,713
15 ,04 7
32,041
216
N orfolk
2,64 0
508
1,690
3,551
59 3
2,436
16
N orth W ales
2,40 0
352
1,170
2,459
29 7
1,686
11
N orth Yorkshire
2,40 0
352
1,170
2,459
59 3
1,686
11
N ortham ptonshire
2,64 0
704
2,339
4,917
1 ,48 3
3,373
23
N orthum bria
9,84 0
743
2,469
5,190
29 7
3,560
24
N ottingha m shire
5,04 0
508
1,690
3,551
1 ,03 8
2,436
16
S outh W a les
6,24 0
665
2,209
4,644
89 0
3,185
21
S outh Yorksh ire
7,20 0
430
1,430
3,005
1 ,26 0
2,061
14
S ta ffordshire
3,60 0
900
2,989
6,283
1 ,55 7
4,310
29
S uffolk
1,92 0
156
520
1,093
7 4
749
5
S urrey
2,40 0
430
1,430
3,005
74 1
2,061
14
S ussex
4,56 0
782
2,599
5,464
81 5
3,747
25
T ham es V alley
7,20 0
1,291
4,289
9,015
2 ,52 0
6,183
42
W arw ickshire
1,92 0
235
780
1,639
22 2
1,124
8
W est M ercia
3,84 0
391
1,300
2,732
59 3
1,874
13
W est M idlands
1 5,84 0
1,682
5,588
11,747
1 ,55 7
8,057
54
W est Yorkshire
1 0,56 0
2,542
8,448
17,756
6 ,96 8
12,179
82
W iltshire
1,68 0
274
910
1,912
7 4
1,312
9
A berde en
1 4
D un dee
1 4
Includ ed below
Included be low
E dinb urgh
1 9
G lasgow
1 4,11 2
S cottish F ingerprin t
S ervice
2,542
8,448
17,756
4 ,52 2
12,179
82
B ritish T ranspo rt
P olice
1,44 0
117
390
820
0
562
4
H M C ustom s &
E xcise
2 40
39
130
273
0
187
1
N ation al Ide ntification
S ervice
8,40 0
508
1,690
3,551
89 0
2,436
16
N ation al C rim e
S quad
2 40
0
0
0
0
0
0
S pecialist C rim es
B ureau
0
0
0
0
0
0
0
T O T AL
24 0,00 0
39 ,11 5
1 29 ,9 63
27 3 ,17 5
7 4,3 47
18 7,37 4
1 ,26 2
Figure 5.2-1 Bureau Maximum Monthly Workload
Schedule F, Service Level Requirements
93
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
The figure above shows the Bureau Monthly Maximum Ten Print and Mark workloads based on each
Bureau’s proportion of the National Ten Print and Mark workload at 2004. As stated above, these
figures will be refreshed annually. The workload is then distributed according to search type: non-
verified Live ID, Print-to-Print, Print-to-Mark, Mark-to-Print, and Mark-to-Mark. Note that
engineering margins have been built into the IDENT1 design to accommodate minor variations of the
distribution of the total search load between the search types from that specified in DOR Table 2.18.
Bureau workload is based on a set of assumptions presented in Figure 5.2-2. These assumptions are
derived from the DOR and the NAFIS SLA, were used in the design of IDENT1 and establish the
operational limits of the system. The assumptions are valid at TOR. At FOC, the FAST priority will
be eliminated and its allocated percentages combined with those of the NORMAL priority. In
addition, the geographically based search scope will be adjusted to meet the Authority’s database
penetration requirements as stated in the DOR (Table titled “Authority Indicative Response Times for
Searches of the Unified Collection”)
TEN PRINT PROCESSING
Quoted CRO Image Retrievals
69%
Print-to-Print AFR Searches
o % URGENT Submissions
10.90%
o % FAST Submissions
22.20%
o % NORMAL Submissions
66.90%
Print-to-Mark AFR Searches
o % URGENT Submissions
2.60%
o % FAST Submissions
32.40%
o % NORMAL Submissions
65.00%
MARK PROCESSING
Mark-to-Print AFR Searches
o % URGENT Submissions
2.60%
o % FAST Submissions
32.40%
o % NORMAL Submissions
65.00%
M-M AFR Searches
o % URGENT Submissions
2.60%
o % FAST Submissions
32.40%
o % NORMAL Submissions
65.00%
SEARCH SCOPE
P-M, M-P, and M-M Searches
o National
5.00%
o Regional
50.00%
o Local
45.00%
Figure 5.2-2 National Work Profiles
This SLR has used this set of assumptions to estimate the workloads expected during normal
operations. Assumptions 1 through 12, 14 and 15 apply only from TOR until FOC.
(1)
Quoted CRO Image Retrieval quantities will be at least 60% of the total Ten Prints received
and processed by the Police Forces. (derived from DOR (Inf-NF-640) “Of the Print Sets
received, current statistics indicate that approximately 69% are Recidivists, with an existing
Schedule F, Service Level Requirements
94
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
criminal record, and 31% have no prior entry on the existing system.”)
(2)
Print-to-Print Urgent priority AFR search quantities will be no more than 10.90% of all Ten
Prints submitted for a Print-to-Print AFR search.
(3)
The combination of Print-to-Print Urgent and Fast priority AFR search quantities will be no
more than 33.10% of all Ten Prints submitted for a Print-to-Print AFR search.
(4)
Print-to-Mark Urgent priority AFR search quantities will be no more than 2.60% of all Ten
Prints submitted for a Print-To-Mark AFR search.
(5)
The combination of Print-To-Mark Urgent and Fast priority AFR search quantities will be no
more than 35.00% of all Ten Prints submitted for a Print-to-Mark AFR search.
(6)
Mark-to-Print Urgent priority AFR search quantities will be no more than 2.60% of all Marks
submitted for a Mark-to-Print AFR search.
(7)
The combination of Mark-to-Print Urgent and Fast priority AFR search quantities will be no
more than 35.00% of all Marks submitted for a Mark-to-Print AFR search.
(8)
Mark-to-Mark Urgent priority AFR search quantities will be no more than 2.60% of all Marks
submitted for a Mark-to-Mark AFR search.
(9)
The combination of Mark-to-Mark Urgent and Fast priority AFR search quantities will be no
more than 35.00% of all Marks submitted for a Mark-to-Mark AFR search.
(10)
National search scope quantities will be no more than 5.00% of each priority search type
within Print-to-Mark, Mark-to-Print, and Mark-to-Mark AFR searches.
(11)
Regional level search scope quantities will be no more than 50% of each priority search type
within Print-to-Mark, Mark-to-Print, and Mark-to-Mark AFR searches.
(a)
A local search is defined as a search of the ‘home’ force area only.
(b)
A regional search is defined as a search involving a minimum of one station
and a maximum of five police force areas. This may include the ‘home’
force but other forces selected do not have to be contiguous with it.
(c)
A national search is defined as a search of the full unified database (i.e., all
forces)
(12)
Local level search scope quantities will be at least 45% of each priority search type within
Print-to-Mark, Mark-to-Print, and Mark-to-Mark AFR searches.
(13)
All Operational Response searches will be processed as Normal priority.
(14)
British Transport Police searches may be 100% National searches rather than the percentages
defined above.
(15)
Customs and Excise Service searches may be 100% National searches rather than the
percentages defined above.
(16)
All Immigration and Nationality Directorate P-P searches will be processed as Normal
priority.
(17)
All Immigration and Nationality Directorate P-M searches will be processed as Low priority.
Schedule F, Service Level Requirements
95
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
PITO
IDENT1 CONTRACT
Northrop Grumman
Released under FOI in full on 16th July 2009
(18)
The workload from the Immigration and Nationality Directorate will be 10% CRO retrievals
and up to 90% Print-to-Print searches.
Schedule F, Service Level Requirements
96
Signature Version
Part 2 Operations Service Level
Released under FOI in full on 16th July 2009
