Forecasting the prison population

Abstract

This paper describes the development of a new methodology tor making long ' term projections of the Prison Population. The mn ethodologv is based on a flow model of offenders in and oui of prison, allowing the effect of changes in court sentencing policy to be predicted. The heart of the methodology is a new theory of offending, al/owing the number of first time offenders to be predicted and also the recidivism ot those released from prison.-00000-In 1996 the Operational Research Unit of the Home Office was asked to develop a new 'Long Term' methodology to forecast the Prison Population in England and Wales. The requirement was to be able to predict the average population (broken down by age, sex, and type of offence) in any year up to around five years in advance.The purpose of these long term forecasts is to inform the Prison Agency's programme of estate management (cg how many new prisons to build, how many to refurbish, how many to close). The existing methods of projection were based on regression and time series models. The problem with such an approach is that as well as projecting changes in the nature of criminal offending, one must also predict changes in sentencing policy by the courts. Since 1993 there as been a dramatic increase (10 to 15%), year upon year, in the use of custodial (prison) sentences by the courts. It is difficult to see how a time series or regression approach can be helpful in such circumstances. Instead it was decided to investigate the possibility of building a 'flow' model which could be used to test the consequences of various policy scenarios (cg reduce the use of custody by 10% for non-violent crime or 'three strikes and you're out'). The final model was Tizo ar/ide remains Coma Gopvriht OR Insight Vol II Issue / -January-March /998 3 intended to be easily used by policy makers to test the results of various policy options.As we will see below, the construction of a flow model requires some understanding of the behaviour of offenders when confronted with the Criminal Justice System. This has resulted in the development of a novel but remarkably accurate theory of offending which will be described in this paper.Together, the 'flow model' and the theory of offending make exceptionally accurate predictions' of the prison population for years up to 1996 and, given accurate scenarios, can be expected to make similarly good future predictions.We describe below a summary of the main features of our model which is near completion. We hope to publish a more detailed description in the near future. A flow modelOne of the reasons for using a flow model is the inherent stability of the prison population over a period of a few years. This is partly due to the greater contributions of those with long sentences to the population (generally, one 'two year' sentence has the same contribution as two 'one year' sentences). One has good information about the current prison population and in particular about long sentence prisoners. As these make a large contribution to the total, they make up ...

Model development in discrete-event simulation and system dynamics: An empirical study of expert modellers

“…Both approaches have been previously used for modelling the prison population system. DES models of the prison population have been developed by Kwak et al (1984), Cox et al (1978), Korporaal et al (2000), and SD models have been developed in Bard (1978), andMcKelvie et al (2007), while the UK prison population model by Grove et al (1998) is a flow model analogous to an SD model.…”

Patient characteristics related to length of stay in Dutch forensic psychiatric care

“…Several factors can have an impact on LOS, such as a country's jurisdiction, changes in criminal justice policy (Grove, MacLeod, & Godfrey, 1998), changes in the organization of a country's mental health system (Sharma et al, 2015), and patient characteristics (Davoren et al, 2015;Dumont et al, 2012;Moran, Fragala, Wise, & Novak, 1999;Shah et al, 2011).…”

Exploring the model development process in discrete-event simulation: insights from six expert modellers

“…The modelling exercise used as the stimulus for inducing modellers' thoughts asked expert modellers to build models of the UK prison population, based on a study reported by Grove et al (1998). The UK prison population problem was considered suitable because it is sufficiently self-contained to enable the development of a simulation model in a limited time period of around 60 minutes.…”