An Interactive Workload and Risk Balancing Model and Decision Support System for Probationer Assignment

The purpose of probation is to protect public safety by deploying personnel to supervise individuals assigned to probation by the court. This paper presents a model for allocating probation officer resources that combines a statistical model for probationer risk classification and an integer, nonlinear multi-criteria programming resource allocation model for workload balancing. Data were obtained from the Florida Department of Corrections, which is implementing the model. The classification model identifies the expected likelihood of failure during discrete periods of supervision. These likelihoods are used to assign individuals to supervision levels. The proportion of probationers in each risk category is an input to the allocation model where decision variables are number of visits to probationers by type of visit and risk classification. Model constraints include performance characteristics, service time, public safety measures, and reduction in probationer failures. Officer contacts are allocated to minimize probationer failures and time and budget overruns.