1. Introduction and Overview. 2. Detecting Influential Observations and Outliers. 3. Detecting and Assessing Collinearity. 4. Applications and Remedies. 5. Research Issues and Directions for Extensions. Bibliography. Author Index. Subject Index.

Regression Diagnostics: Identifying Influential Data and Sources of Collinearity

OR/MS research in disaster operations management

Criminal careers have long occupied the imaginations of criminologists. Since the 1986 publication of the National Academy of Sciences report on criminal careers and career criminals, a variety of theoretical, empirical, and policy issues have surfaced. Data on key criminal career dimensions of prevalence, frequency, specialization, and desistance have raised theoretical questions regarding the patterning of criminal activity over the life course. Recent research has identified important methodological issues, including the relationship between past and future criminal activity, and potential explanations for this relationship: state dependence and persistent heterogeneity. Advanced statistical techniques have been developed to address these challenges. Criminal career research has identified important policy issues such as individual prediction of offending frequency and career duration, and has shifted the focus toward the interplay between risk and protective factors.

The Criminal Career Paradigm

https://www.rand.org/content/dam/rand/pubs/reports/2006/R1238.pdf

A hypercube queuing model for facility location and redistricting in urban emergency services

The evolution of the intellectual structure of operations management—1980–2006: A citation/co-citation analysis

This paper gives a model in which two urban emergency service units such as fire engines or ambulances cooperate in responding to alarms or calls from the public in a specified region of a city. Given the home locations of the units and the spatial distribution of alarm rates, it is possible to specify which unit should respond to each alarm by defining a response area for each unit. The average response time to alarms and the workload of each unit are calculated as functions of the boundary that separates their response areas. The boundaries that minimize average response time and the ones that equalize workload are determined. Some boundaries can be dominated, in the sense that another boundary improves both workload balance and response time. The set of undominated boundaries is found.

/pdf/response-areas-for-two-emergency-units-1kfxo72pxy.pdf

Response Areas for Two Emergency Units

An urban emergency service system provides mobile units vehicles to respond to requests for service which can occur at any time and any place throughout a city. This paper describes the common characteristics and operational problems of these systems and surveys the various methods, both traditional and recently developed, which may be used for allocating their units. Aspects of allocation policy discussed include 1 determining the number of units to have on duty, 2 locating the units, 3 designing their response areas or patrol areas, 4 relocating units, and 5 planning preventive-patrol patterns for police cars.

Typical policy changes which may be suggested by the use of quantitative allocation models include selective queuing of low priority calls, varying the number of units on duty and their locations by time of day, dispatching units other than the closest ones to certain incidents, relocating units as unavailabilities begin to develop, and assigning police cars to overlapping patrol sectors. As a result of making such changes, it is often possible to reduce queuing and travel time delays, improve the balance of workload among units, and enhance the amount of preventive patrol where needed.

/pdf/methods-for-allocating-urban-emergency-units-a-survey-5ga9zw2nbo.pdf

Methods for Allocating Urban Emergency Units: A Survey

A computer program has been designed for specifying the number of police patrol cars that should be on duty in each geographical command of a city at various times of day on each day of the week. It incorporates, by user option, nearly all the desirable features of earlier allocation programs, together with several improvements. It calculates performance statistics for the current allocation of patrol cars or any allocation proposed by the user. In addition, it has two prescriptive capabilities: 1 determining the minimum number of patrol cars needed during each tour in each command to meet specified constraints on performance measures, and 2 allocating a specified total number of car-hours by time and/or geography so as to optimize one of several available objective functions. The main technical innovation in the model is that it allows one tour in each day to overlay two other tours. A heuristic algorithm allocates car-hours when there is such an overlay tour; it is optimal when the overlay tour has the same duration as the tours it overlays.

A Patrol Car Allocation Model: Capabilities and Algorithms

Before designing a computer program for allocating police patrol cars by time and geography, a review was undertaken of previously existing programs of this type. Nearly all of the programs calculated queuing statistics for the collection of patrol cars by assuming a steady-state system with calls for service arriving within priority levels according to Poisson processes and having independent, identical, exponentially distributed service times. Unavailabilities of patrol cars for reasons other than calls for service were handled in the models either by artificially increasing the arrival rate of calls or by assuming that the number of servers is smaller than the number of patrol cars. Some programs calculated additional performance measures such as travel times and preventive patrol frequencies. All the programs had the capabilities to describe performance statistics for an allocation proposed by the user, but they differed in their capabilities to prescribe desirable allocations. None of the programs had achieved general acceptance because each had virtues and inadequacies not present in the others.

Jan M. Chaiken

Papers

Response Areas for Two Emergency Units

Methods for Allocating Urban Emergency Units: A Survey

A Patrol Car Allocation Model: Capabilities and Algorithms

A Patrol Car Allocation Model: Background

Indicators of Crime and Criminal Justice: Quantitative Studies.