Design and estimation for small sample sensitivity testing

In sensitivity testing for the Department of Defense, the high cost of experimental units necessitates the use of small sample sizes and accentuates the importance of design. This article compares five data collection-estimation procedures. Four of these are modifications of the Robbins-Monro method, and the other is the Langlie. The simulation study is designed as a factorial experiment with response function, sample size, initial design point, gate width, and noise as factors. The estimated V₅₀ and its MSE are the responses compared to assess the small sample behavior of each method. Although there is no single clear-cut winner, the Delayed Robbins-Monro (DRM) with maximum likelihood estimation and the Estimated Quantal Response Curve (Wu [21]) are shown to perform well over a broad variety of conditions.     

Ordering policies for an inventory system with three supply modes

This article explores ordering policies for inventory systems with three supply modes. This model is particularly interesting because the optimal ordering decision needs to balance the inventory and purchase costs, as well as the costs for earlier and later periods. The latter cost trade-off is present only in inventory systems with three or more supply modes. Therefore, the result not only offers guidelines for the operation of the concerned inventory systems, but also provides valuable insight into the complex cost trade-offs when more supply modes are available. We assume that the difference between the lead times is one period, and the inventory holding and shortage costs are linear. We analyze two cases and obtain the structure of the optimal ordering policy. Moreover, in the first case, explicit formulas are derived to calculate the optimal order-up-to levels. In the second case, although the optimal order-up-to levels are functions of the initial inventory state and are not obtained in closed form, their properties are discussed. We also develop heuristic ordering policies based on the news-vendor model. Our numerical experiments suggest that the heuristic policies perform reasonably well. © 1996 John Wiley & Sons, Inc.     

General solution to many-on-many heterogeneous stochastic combat

Stochastic combat models are more realistic than either deterministic or exponential models. Stochastic combat models have been solved analytically only for small combat sizes. It is very difficult, if not impossible, to extend previous solution techniques to larger-scale combat. This research provides the solution for many-on-many heterogeneous stochastic combat with any break points. Furthermore, every stage in stochastic combat is clearly defined and associated aiming and killing probabilities are calculated. © 1996 John Wiley & Sons, Inc.     

The advantage of deeper pockets in Silverman's game on intervals

Silverman's game on (1, B) × (1, B) was analyzed by R. J. Evans, who showed that optimal strategies exist (and found them) only on a set of measure zero in the parameter plane. We examine the corresponding game on (1, B) × (1, D) with D > B, and show that optimal strategies exist in about half the parameter plane. Optimal strategies and game value are obtained explicitly. © 1995 John Wiley & Sons, Inc.     

An exact aggregation/disaggregation algorithm for large scale markov chains

Many Markov chain models have very large state spaces, making the computation of stationary probabilities very difficult. Often the structure and numerical properties of the Markov chain allows for more efficient computation through state aggregation and disaggregation. In this article we develop an efficient exact single pass aggregation/disaggregation algorithm which exploits structural properties of large finite irreducible mandatory set decomposable Markov chains. The required property of being of mandatory set decomposable structure is a generalization of several other Markov chain structures for which exact aggregation/disaggregation algorithms exist. © 1995 John Wiley & Sons, Inc.     

Expressions for item fill rates in periodic inventory systems

We examine the problem of estimating the item fill rate in a periodic inventory system. We show that the traditional expressions for line item fill rate, found in many operations management textbooks, perform well for high fill rates (above 90%), but they consistently underestimate the true fill rate. The problem of underestimation becomes significant as the fill rate falls below 90% and is greatly amplified in cases with very low fill rates (below 50%). We review other more accurate expressions for fill rate, discussing their relative merits. We then develop an exact fill rate expression that is robust for both high and low fill rates. We compare the new expression to others found in the literature via an extensive set of simulation experiments using data that reflect actual inventory systems found at Hewlett-Packard. We also examine the robustness of the expressions to violations in the underlying assumptions. Finally, we develop an alternative fill rate expression that is robust for cases of high demand variability where product returns are allowed. © 1995 John Wiley & Sons, Inc.     

Should we be using cost of capital in calculating holding costs of consumable inventory?

The DOD directs the usage of 10% of item cost as the cost of capital in the calculation of inventory holding costs. This 10% cost is not totally justified and a complete review must be accomplished to bring this factor to a meaningful and more useful value. The current logic supporting a 10% cost of capital results in a continuing perturbation which forces the Air Force to operate in a less than efficient mode when using the economic order quantity for consumable purchases.     

A model for use in the rationing of inventory during lead time

While the traditional solution to the problem of meeting stochastically variable demands for inventory during procurement lead time is through the use of some level of safety stock, several authors have suggested that a decision be made to employ some form of rationing so as to protect certain classes of demands against stockout by restricting issues to other classes. Nahmias and Demmy [10] derived an approximate continuous review model of systems with two demand classes which would permit an inventory manager to calculate the expected fill rates per order cycle for high-priority, low-priority, and total system demands for a variety of parameters. The manager would then choose the rationing policy that most closely approximated his fill-rate objectives. This article describes a periodic review model that permits the manager to establish a discrete time rationing policy during lead time by prescribing a desired service level for high-priority demands. The reserve levels necessary to meet this level of service can then be calculated based upon the assumed probability distributions of high- and low-priority demands over lead time. The derived reserve levels vary with the amount of lead time remaining. Simulation tests of the model indicate they are more effective than the single reserve level policy studied by Nahmias and Demmy.     

The reliability of a series system of repairable subsystems: A bayesian approach

Consider the problem of estimating the reliability of a series system of (possibly) repairable subsystems when test data and historical information are available at the component, subsystem, and system levels. Such a problem is well suited to a Bayesian approach. Martz, Waller, and Fickas [Technometrics, 30 , 143–154 (1988)] presented a Bayesian procedure that accommodates pass/fail (binomial) data at any level. However, other types of test data are often available, including (a) lifetimes of nonrepayable components, and (b) repair histories for repairable subsystems. In this article we describe a new Bayesian procedure that accommodates pass/fail, life, and repair data at any level. We assume a Weibull model for the life data, a censored Weibull model for the pass/fail data, and a power-law process model for the repair data. Consequently, the test data at each level can be represented by a two-parameter likelihood function of a certain form, and historical information can be expressed using a conjugate family of prior distributions. We discuss computational issues, and use the procedure to analyze the reliability of a vehicle system. © 1994 John Wiley & Sons, Inc.     

Mean residual life of coherent systems consisting of multiple types of dependent components

Mean residual life is a useful dynamic characteristic to study reliability of a system. It has been widely considered in the literature not only for single unit systems but also for coherent systems. This article is concerned with the study of mean residual life for a coherent system that consists of multiple types of dependent components. In particular, the survival signature based generalized mixture representation is obtained for the survival function of a coherent system and it is used to evaluate the mean residual life function. Furthermore, two mean residual life functions under different conditional events on components’ lifetimes are also defined and studied.