WITNESS simulation software has been used by the Norfolk Constabulary in a far-reaching programme to help improve its Control Room operations, monitor the efficiency of vehicle response to incidents, and optimise the use of custody suites county-wide.

Using WITNESS Norfolk police were able to thoroughly examine and verify the effectiveness of different solutions that had been proposed. Inspector Alan Brighten, from the Project Support Unit of the Norfolk Constabulary's Corporate Development Department, says "WITNESS has proved an extremely valuable tool in allowing us to evaluate the impact of change and to prove its effectiveness."

In terms of geographical area covered, of the 43 Forces in England and Wales, Norfolk Constabulary is the ninth largest, but with a strength of 1,432 has the third lowest number of officers per head of population. There are more than 120 incidents-per-officer per year - around 750 a day - and almost 70,000 999 calls, in addition to more than a million non-emergency calls.

"We are under more pressure than ever before to improve performance" reports Inspector Brighten. "In addition to the key performance indicators we are compelled to provide to the Home Office, we are constantly under scrutiny from HM inspectors of Constabulary, the Police Authority, the Audit Commission - and of course the general public."

Although the Home Secretary Jack Straw has instigated a move towards nationwide standardisation, today each force still measures performance in different ways; benchmarking is as yet impossible.

The Control Room

The Control Room, where 999 calls are received, is usually the first point of contact with the public needing police services, and effective working practices are essential to provide the best service possible. High speed response is paramount. The performance indicators that Norfolk Police sets itself were unequivocal: 88 per cent of incoming calls should be answered in less than ten seconds; response to emergency calls in urban areas should take less than ten minutes and for rural areas, less than 15 minutes.

In 1996 Norfolk Police implemented a Computer Aided Despatch, a new computerised call logging system, to replace a paper-based operation with a hand-written message for every call, and brought together six separate Control Rooms and switchboards into one centralised facility.

The transition was problematic, exacerbated by the heavy increase in 999 calls - often as a result of numerous calls from mobile phones reporting the same incident, which caused 'spikes' in incoming 999 calls.

As a result, the effectiveness of response to 999 calls actually fell with the new system. In 1995/96 Norfolk Police were achieving their target of 88 per cent of calls answered inside ten seconds; in 1996/97, with the new computerised/centralised system, this fell to 73 per cent. In some months it fell as low as 50 per cent. This was not acceptable. It was at this stage that consultants hired to identify the problems suggested the use of WITNESS simulation software to model Control Room operations.

The accuracy of the modelling was quick to prove itself. For example, WITNESS came up with the figure of 25 calls on average missing the response-time target each day; the actual, 'real life', number was between 25 and 26. 'Results like these indicated that we could rely on the accuracy of WITNESS to examine and verify any changes we brought about,' says Inspector Brighten.

WITNESS subsequently showed that the idea of identifying dedicated 999 call-takers would solve much of the problem, and a decision was taken to implement the system. As well as verifying the effectiveness of the new operating procedures, WITNESS was used to prove the case for more staff, new equipment and a new shift system-recommendation that have now all been adopted.

The improvements since the WITNESS-verified Control Room procedures have been put in place are impressive. Now almost 95 per cent of 999 calls are picked up inside the ten-second target.

Vehicle Response System

Following the successful use of WITNESS in helping to improve the efficiency of the Control Room operation, Norfolk Police used it in a drive to enhance vehicle response times. "Again it was a question of measurement" Inspector Brighten explains. "We had no data on how much time police vehicles were spending travelling to incidents. Ours is a large geographical area, with a seasonally-variable population, and we had no system to measure the effectiveness of changes to the location and availability of units. We needed a better understanding of the logistics of incident response, and the effects of deploying extra staff in busy periods. So we built a model using WITNESS."

Information fed into the WITNESS model included details of actual incidents, their location, the time it took to send out a police unit, the location of the nearest unit, the time it spent travelling to the scene and the time spent at the scene. This model enables Norfolk Police to measure almost all elements of vehicle response, and allocate resources accordingly. It has identified times for each stage of the incident call, taking into account variables such as traffic density, type of incident and possible travel speed.

The drawback here is that the system has to work on averages; not all police work is quantifiable and unit availability is indeterminable. Traffic cars work to routes, not geographical areas and there is no standard formula for measuring the seriousness of incidents or whether they lead to re-deployments or require more than one unit. Also officers may not be able to report back, on arrival at an incident.

"But what clarify emerges from the modelling process is a better understanding of staffing requirements in busy periods, as well as the workloads placed on individual units." says Inspector Brighten. "We now have real data available, from our Computer Aided Despatch System, to feed the model and a highly accurate Geo Base to pinpoint locations down to ten metres, The effect on the response time of moving vehicle locations is measurable, as is the effect of shift times for the units, and we can measure the requirement for extra staff to cope with the seasonal variations in the area. We can also provide an effective demonstration, to Police or public authorities, of the reasons for unit deployment. The model is not exact, but it will effectively demonstrate variations in performance."

The use of WITNESS is currently being extended into a Force Model to measure weighting by Divisions (this is for a review currently under way), and a Division Model to assess the effects of changes to locations and shifts of various units. This new model is being developed to deal specifically with emergency incidents occurring at the same time as those needing attendance within two hours, including the facility to drop non-emergency messages and re-deploy to handle the emergency.

The Custody Model

As well as operating in the front line, WITNESS is helping Norfolk police in the back room. A simulation model is being used to assess changes in the way custody suites are being used.

Inspector Brighten explains the reasons behind the move: "The Police and Criminal Evidence Act puts clear obligations on the Service in relation to the treatment of prisoners, particularly in the light of recent deaths-in-custody investigations and reports by the Police Complaints Authority. As well as this ´duty of care´ to our prisoners we are rationalising the use of custody suites across the County, and investigating the possibility of cross-force boundary cooperation. Along with the specialised resources and substantial outlay involved in compulsory cell modifications and the installation of, for example, CCTV systems, all this adds up to change which needs careful evaluation and management. "

"Since WITNESS had proved so useful in the change management involved in the implementation of the new Control Room, and the evaluation of the Vehicle Response System, we are using it again to assess the impact of changes to enable us to arrive at the best possible use of custody suites."

The Custody Model will examine elements such as cell capacity requirements, the number and sex of custody staff, and their shift options. The Model will also account for different types of prisoner - males, females, juveniles, drunks, those acting violently, for example - and their custodial requirements.

Painless experimentation

"WITNESS does not in itself present solutions" Inspector Brighten states. "But it gives us an invaluable tool to test the effectiveness of the solutions that we propose, and to examine, evaluate and verify a whole range of 'what if?' options. The beauty of WITNESS is that it enables painless experimentation."

Norfolk Constabulary is now considering possible future applications of WITNESS inside the organisation to enhance the service provided to the public of Norfolk. There is also a place for WITNESS in helping with the design and overall working practices at the Force's new headquarters soon to be established in Wymondham.