System comparison procedures for automatic target recognition systems

Estimating the performance of an automatic target recognition (ATR) system in terms of its probability of successfully identifying a target involves extensive image collection and processing, which can be very time‐consuming and expensive. Therefore, we investigate the Wald sequential test for the difference in two proportions as a sample size‐reducing alternative to ranking and selection procedures and confidence intervals. An analysis of the test parameters leads to a practical methodology for implementing the Wald test for fairly comparing two systems, based on specific experimental goals. The test is also extended with the modified, sequentially rejective Bonferroni procedure for the multiple pairwise comparison of more than two systems. Two sampling schemes for different experimental goals are also discussed. The test methodology is applied to actual data to compare different configurations of a specific ATR system, with the results demonstrating that the modified Wald sequential procedure is a useful alternative to comparing proportions with confidence intervals, particularly when data are expensive. © 1999 John Wiley & Sons, Inc. Naval Research Logistics 46: 357–371, 1999     

The residual life of the renewal process: A simple algorithm

We develop a simple algorithm, which does not require convolutions, for computing the distribution of the residual life when the renewal process is discrete. We also analyze the algorithm for the particular case of lattice distributions, and we show how it can apply to an inventory problem. © 1999 John Wiley & Sons, Inc. Naval Research Logistics 46: 435–443, 1999     

Rebellion,Repression and Welfare

I develop a dynamic model of social conflict whereby manifest grievances of the poor generate the incentive of taking over political power violently. Rebellion can be an equilibrium outcome depending on the level of preexisting inequality between the poor and the ruling elite, the relative military capabilities of the two groups and the destructiveness of conflict. Once a technology of repression is introduced, widespread fear reduces the parameter space for which rebellion is an equilibrium outcome. However, I show that repression-driven peace comes at a cost as it produces a welfare loss to society.     

Book reviews

John Lawrence Tone, The Fatal Knot: The Guerrilla War in Navarre and the Defeat of Napoleon in Spain. Chapel Hill and London; The University of North Carolina Press 1994. Pp.vii + 239, 1 map, biblio., index. $34.95 (cloth). ISBN 0–8078–2169–1.

Robert Holland (ed), Emergencies and Disorder in the European Empires after 1945. London: Frank Cass, 1994. Pp.x + 256; index. £32 (cloth); £15 (paper). ISBN 0–7146–4516–8 and 4109 X

Deborah L. Norden, Military Rebellion in Argentina: Between Coups and Consolidation. Lincoln, NE: University of Nebraska Press, 1996. Pp.242, index, $35/$17.50 (paper). ISBN 0–8032–8369–5.

William M. Minter, Apartheid's Contras: An Inquiry into the Roots of War in Angola and Mozambique. London and Atlantic Highlands, NJ: Zed Books Ltd; Johannesburg: Witwatersrand University Press, 1994. Pp. xii +308, 3 maps, biblio., index. $69.95/£39.95 (cloth); $29.95/£16.95(paper). ISBN 1–85649–266–4 and 266–4.     

On estimating survival for reliability models based on age-related stochastic comparisons

A new class of nonparametric reliability models is introduced and studied. A distribution is said to be better at age s than at age t (sBt) if the residual lifetime at age s is stochastically greater than or equal to the residual lifetime at age t. Applications to various forms of replacement policies, including the cannibalization of failed systems, are noted. For fixed s < t, the problem of estimating a survival curve assumed to belong to the sBt class is addressed using recursive methods. An sBt estimator is derived in closed form, and its uniform strong consistency at an optimal rate of convergence is demonstrated. A simulation study strongly supports the claim that the sBt estimator tends to outperform the empirical survivor function in small- and moderate-size samples. © 1993 John Wiley & Sons, Inc.     

Location-allocation for distribution to a uniform demand with transshipments

This article analyzes the location-allocation problem for distribution from a single fixed origin via transshipment terminals to a continuous uniformly distributed demand. Distribution through terminals concentrates flows on the origin-to-terminal links and transportation economies of scale encourage the use of larger vehicles. Analytical expressions are derived for the optimal terminal locations, the optimal allocation of destinations to terminals, and the optimal transportation cost. Continuous analytic models assume either an allocation, by partitioning the service region into sectors, or terminal locations. This is unlikely to produce an optimal distribution system. The optimal cost is compared to the cost for suboptimal location-allocation combinations. Results indicate that the location decision is not too important if destinations are allocated optimally and that allocation to the nearest terminal may be poor, even with optimal locations. © 1992 John Wiley & Sons, Inc.     

Algorithms for the design of finite-capacity service units

Finite-capacity queues arise naturally in many practical situations, notably in communications and manufacturing engineering. In this article, the matrix formalism of probability distributions of phase type is used to develop fast algorithms to compute various steady-state distributions for the finite-capacity PH/PH/1 queue. This algorithm is an important ingredient in solving nontraditional but significant design problems. Some of these are described as illustrative examples.