The model is a system of three equations. The first is normally referred to as a supply of offenses function and suggests that the crime rate is functionally related to the probability of arrest and social characteristics, among which economic variables (incidence of unemployment and poverty) are likely to be very important. The second equation posits that the arrest rate is a function of inputs, the volume of crime, and the stability of a neighborhood. The third equation estimates the factors and parameters behind the present system of allocation of police inputs. Model parameters are shown to be an aid to decisionmakers if they are pursuing a given criterion for performance, and the impact of the adoption of each of a range of 'rational' criteria for the distribution and incidence of robbery in New York City is examined. The model was tested with data for New York City for 1973, using as the endogenous variables the reported rate of robberies per head, the reported clearance rate for robbery, and the total precinct patrol officer strength, adjusted for nondiscretional manpower. The ways in which various patterns of manpower deployment can be used to affect the crime rate are examined through the use of the model. Limitations of the analysis are discussed. Tables and footnotes are provided.