Search

The matching identification problem (MIP) is a combinatoric search problem related to the fields of learning from examples, boolean functions, and knowledge acquisition. The MIP involves identifying a single “goal” item from a large set of items. Because there is commonly a cost associated with evaluating each guess, the goal item should be identified in as few guesses as possible. As in most search problems, the items have a similar structure, which allows an evaluation of each guessed item. In other words, each guessed item elicits partial information about the goal item, i.e. how similar the guess is to the goal. With this information the goal is more quickly identified. The unordered MIP has been studied by Mehrez and Steinberg (ORSA J. Comput. 7 (1995) 211) in which they proposed two different types of algorithms. The purpose of the present paper is to suggest an improved Spanning Heuristic algorithm. Its improvement increases as the problem size increases. Further results and comparisons are derived for the unordered and ordered cases. This research shows that when the search space is very large, it is better to inquire from items that are known not to be the goal (they have been ruled out by previous guesses), for the purpose of acquiring more information about the goal. As the search space is narrowed, it is better to guess items that have not been ruled out.

For economic and other reasons, distribution networks are often constructed in hierarchies, where, due to high costs, high-level distribution channels are built in straight lines from which low-level channel branch. Branching facilities, too, may incur high costs and thus, their number, locations along the lines, and allocation of destinations to them are important components in system design, which can be accomplished by solving a series of capacitated location-allocation problems on a line. This problem is studied here and its properties are discussed, and appropriate mathematical optimization models formulated and their properties and complexity are also considered. Heuristic solution schemes and lower bounds on objective values are proposed for the models. The efficiency and effectiveness of these schemes are evaluated and more insight into the models is gained through a numerical study. The major conclusion is that the hierarchical structure considered is highly restrictive and imposes high costs. Therefore, its use must be properly justified.

This paper analyzes the strategic calculations of an academic program when setting minimal quality requirements for applicants as an issue in planning a signaling strategy under certain structural conditions, such as a rapid growth of a higher education system. We present two versions of a signaling game between an academic program/institution, as the informed player, and a governmental controlling body that has incomplete information about the program's quality. One version describes a situation where there is low sensitivity to quality variations in the system and the second where there is high sensitivity to quality. Analysis shows that governmental controlling bodies can affect the quality of the system by influencing the sensitivity to quality rather than by direct and tight control over academic institutions. Given a certain level of sensitivity, the impact of these bodies on outcomes is minimal so their intervention in the market can be limited as well.

The time-to-market in the presence of a window of opportunity is analyzed using ;a probabilistic model, i.e. a model where the completion time of new product development is a random variable characterized by a gamma distribution. Two cases are considered: the first, a case where the discounted return-on-investment exceeds the return expected from a conservative investment—e.g. investment in bonds—termed ‘the profitable case’; and the second, a case where the discounted return-on-investment just balances the cost of new product development, termed ‘the salvageable case’. The model constructed is focused on the financial aspects of new product development. It allows a decision-maker to monitor, as well as terminate, a project based on its expected value (at any time prior to completion) by computing the mean time-to-market that provides profit, investment salvage, or loss. The mean time-to-market computed by the model may be compared with that estimated by the technology development team for decision-making purposes. Finally, in the presence of a window of opportunity and for the specific cases analyzed, we recommend to always keep the expenditure rate lower than the expected return rate. This will provide the decision-maker a salvageable exit opportunity if project termination is decided. Copyright © 2002 John Wiley & Sons, Ltd.

Inventory management involves determination of shortage policy. It specifies the conditions for losing or backordering a demand. Alternative policies include pure backorder, pure lost sales, and partial backorder (using a single backorder control limit). When the backorder-cost is time dependent it makes sense to modify the backorder-limit over time. Thus, a new form of partial backorder policy (PB2) with two-segment backorder control limits is introduced. The traditional policies mentioned above, are special cases of PB2. Hence, we provide a unified framework for studying different policies that deal with shortage. The PB2 problem is formulated and solved as a discrete time, stochastic constrained control problem. Its performance is numerically compared with the simpler alternative policies. In some cases its cost savings, versus the best of PB and PL, exceeds 15%, and 7% versus a single backorder limit policy. The economical advantage is significant over a wide range of the problem parameters. Scope and purpose This paper develops an expanded framework for modeling shortages in inventory management. It recognizes that optimal backordering strategy may change over time during an “out-of-stock” period. The paper is motivated by experience in the chemical industry in which, the cost of backordering is highly time related. Inventory managers, in this industry, consider to lose sales initially (once they run out of stock) and begin to backorder demand later as they approach the replenishment time. A two-segment partial backorder (r,Q) model is introduced and solved. Pure backorder (PB), pure lost sales (PL), and partial backorder (using a single backorder limit), are all special cases of the proposed model. The problem is formulated and solved as a discrete time, stochastic constrained control problem. Its performance is numerically compared with the simpler alternative policies. In some cases its cost savings, versus the best of PB and PL, exceeds 15%, and 7% versus a single backorder limit policy. The economical advantage is significant over a wide range of the problem parameters. The partial backorder policy we propose is not only different from those in the literature, but it provides new control flexibility.

The Dead Sea Works Ltd extracts more than 2 million ton of potassium a year from the Dead Sea and ships it by surface and marine transportation to markets around the world. Product mix and market distribution changes have made it necessary to redesign the surface logistics system. This system consists of four sites with a total of more than 300,000 ton of roofed storage capacity. The purpose of the study was to suggest changes in the storage capacity of each product at each of the sites. We have designed a model that expresses the expected annual loss for the firm and have solved it under potential market scenarios and management and technological constraints. Sensitivity analyses were then carried out for several uncertain cost and reliability parameters. Based on the best system configuration for each scenario, alternative configurations were designed in detail and compared under potential market scenarios. With a clear understanding of the risks and expected benefits involved, the decision makers then selected the best configuration. The annual net benefit from implementing the proposed solution is expected to be 1.7 million dollars.

Inventory systems with limited repair capacity are affected by congestion externalities, caused by use of a shared service. There is incompatibility between individual and system optimisation in considering congestion externalities. Three models are described that investigate the congestion effect in a multi-echelon inventory system which has two modes of repair, each with a limited repair capacity. An expanding repair policy employed by the bases in order to choose which repair mode to use is described and compared with different expediting policies related to congestion externalities. The expanding repair policy that considers congestion externalities was found to lead to better system performance measurement than an expanding policy with no congestion. The results of the numerical experiment indicate that the model that ignores congestion externalities—that is, the model that measures each base as an individual—leads to poorer performance measurement for every expediting repair policy, and particularly for the optimal expediting repair policy.

The problem explored was of using two machines, one at a time, to perform a continuous operation, where each machine is subject to different types and lengths of maintenance. The operation periods between maintenance periods can also vary in length. All the periods are deterministic in length. When the last period ends, the first period starts again. The objective is to maximize the portion of time with an operating machine. We conjecture this is an NP problem. This problem extends the study by J. Kreimer and A. Mehrez (European Journal of Operational Research 69 (1993) 50–54) on an n -machine single type of maintenance scheduling (MS) problem. The no-interrupt policy, that was optimal for their problem is not optimal for ours. Properties of the optimal solution and identification of the state transition mechanism lead to a finite, but exponentially bounded, algorithm that constructs cyclic solutions. Heuristic methods are recommended and tested for small-sized problems. They achieve, on an average, more than 99% of the highest objective function value found by the algorithm, and equal that value in 55–80% of the random problems. The most efficient heuristic method consists of the SPT scheduling rule.

This paper provides confidence intervals for the probability of bankruptcy through the control of financial accounting variables. Our analysis differs in several aspects from standard bankruptcy techniques studied in previous studies. This bankruptcy literature generally provides classification techniques, peruses classification accuracy, and produces point estimators of bankruptcy for each firm. Various measures concerned with the confidence intervals are studied to evaluate the risk involved in predicting the probability of bankruptcy; for example, their maximum and minimum lengths, and their maximum lower bound and minimum upper bounds. We show that local minimum and maximum lengths are global. The empirical results illustrate a substantial improvement (reduction) in the length and the minimum upper bound of the confidence intervals at the optimal level of the financial accounting variables, whereas the lengths at the industry averages were significantly lower. The results are robust for three, two, and one year prior to bankruptcy.

Two cases of the discrete time finite horizon technology replacement problem are solved. The first deals with two available machines, one in use and a better one that can be purchased to replace it. The second case considers, in addition, a third machine of future technology that will be available at some random future time. The maintenance level of each used machine is chosen for each period in order to economically control performance deterioration. For solving the model, we first derive an optimal preventive maintenance policy by showing that “bang-bang” (i.e., full or non), non-increasing, maintenance efforts are optimal. Employing the optimal maintenance policy, we reformat the problem and conduct a numerical search in order to derive the replacement policy that will maximize the expected net present profit. Numerical examples demonstrate that management may improve profitability by delaying (but not necessarily foregoing) replacement with an available better machine.

This paper deals with the decision making for resource sharing and scheduling for a class of computer-integrated manufacturing cells. Each such cell is characterized by a set of resources and it can be used for the cyclic production of several products. The production (processing) of each product requires a set of operations with sequence constraints among them. Each operation can be performed using alternative modes. Each mode defines the subset of resources needed for the operation and the timing of their use. Resource setup times are sequence-dependent. The problem is to choose the best mode for each operation and, accordingly, to allocate and schedule resources in order to minimize make-span time. A general (feasible) mixed 0–1 LP model is proposed. A customized Branch and Bound type solution algorithm is presented, which fathoms only feasible solutions and finds all optimum solutions.

Several authors have studied a special form of a combinatorial game called the Matching Identification Problem (MIP). It is also a special case of a knowledge acquisition problem. The problem involves identifying a chosen subset of size K out of N symbols through a series of inquiries and corresponding responses. Previous heuristics and optimal algorithms for minimizing the average inquiries required were proposed for only the unordered MIP, where the order of the K symbols in each subset is irrelevant. In this paper, the ordered MIP is introduced, whose initial solution space is K! times larger, but whose structure allows it to gain more information at each inquiry. Three previously proposed heuristics are extended from the Unordered MIP and applied to the Ordered MIP in a computational study. The purpose is to observe how the availability of extra information can help reduce a solution space more quickly. The results show that in many cases the Ordered MIP can be solved more quickly than the corresponding Unordered MIP. This implies that information can be valuable in reducing a large solution space quickly. Scope and purpose Reducing a large number of possible solutions quickly is a common problem in many areas. Creating knowledge-based expert systems requires acquisition of relevant information from a large set of data. Searching the Internet for a particular company’s home page also requires reducing a large number of possibilities to one. Since there are costs associated with the steps involved in reducing a large number of possibilities, we would like to accomplish this reduction as quickly as possible. The Matching Identification Problem (MIP) is a combinatoric problem that begins with a large number of possible solutions, and through a series of steps, reduces the number of possible solutions to one. In the context of the MIP, this paper proposes that extra information can outweigh an increase in the number of possible solutions. Therefore, information availability can outweigh increased solution space size in the context of the MIP.

We present a taxonomy of the research that includes the relationship between quality and inventory. A literature review of over forty-five books and journal articles provides a basis for understanding the research in this area. Our goal is not to review all variants of this relationship or to describe all of the results, but rather to provide a broad overview of the major problems in the literature. We review optimization, statistical analysis and simulation, viewpoint and policy literature, although the majority of the works deal with optimization research and viewpoint or case studies. This is the first attempt to review this subjective matter.

In this paper we consider a real-time multiserver (such as machine controllers) and multichannel (such as regions and assembly lines) systems involving maintenance. Our objective is to fit these systems into the framework of queueing models and thus to justify the use of the powerful queueing theory analytical methods in the analysis of real-time systems. The main difficulty is that real-time systems by their very nature do not permit queues. To resolve this contradiction we use a dual approach in which we consider jobs as servers and servers as jobs. We adjust the traditional definition of availability for the real-time systems under consideration and show how to compute the system’s availability, when both service and maintenance times are exponentially distributed (birth-and-death process). At this stage we restrict ourselves to a worst case (maximum load regime), which is most typical in high-performance data acquisition and control (production and military) systems.

A major issue in a decision to adopt advanced technologies is the justification process. Many technological developments are evaluated according to tangible criteria while important intangible criteria are overlooked. This study investigated a military need for advanced navigation technology in the Israeli Defense Force. The first and second parts discuss difficulties with current navigation methods and present the global positioning systems technology (GPS) as a solution. The third part presents some difficulties with traditional decision-making techniques as they apply to the decision to adopt GPS technology in a military environment. The fourth part presents an application of the analytic hierarchy process (AHP) for defining tangible and intangible decision criteria, assigning them numerical weights and evaluating GPS alternatives accordingly. In conclusion, the study presents an alternative to traditional decision-making processes and military purchasing procedures of innovative technology (such as the lowest bidder method) by providing an opportunity to incorporate user-defined criteria into the decision-making process. This study proposes an alternative to traditional military and governmental purchasing methods and it should be valued as an alternative decision tool for innovative technology in times of shrinking governmental budgets.

The purpose of this work is to review new diretioons in the theory of nonlinear utility forms and to illustrate their development for OR/MS types of problems. The illustrations are borrowed from strategic information analysis, operations management, and health care management.

Technology policies and their effect on technology transfer from defense to civilian industries in Israel from 1967 to 1995 are explored. Defense technology policy is found to drive the growth of the defense industry and to limit defense conversion to commercialization initiatives developed primarily through intra-organizational technology transfer. The case of the Israel Aircraft Industries (IAI) is presented. Also the development of national technology policy in Israel is traced and found wanting. Civilian and defense technology policies are found to be uncoordinated with each other and together are incapable of producing interorganizational technology transfer so as to significantly effect defense conversion on a wide number of economic actors particularly small firms. To achieve a socially effective defense conversion process (in which technologies are transferred from defense firms to many economic actors), the analysis suggests that a national technology policy should be designed and implemented under Prime Minister's leadership.

Previous research on the newsboy problem is based on the assumption that in case of a shortage, unsatisfied demand is lost. Such an assumption is inappropriate for items that have a close substitute. In this paper, we formulate a two-item newsboy problem with substitutability (TINPS). Upper and lower bounds on the optimal order quantities of the two items are derived. Since analytical solutions to the problem are difficult to obtain, a Monte Carlo simulation is used to identify the optimal solution to the TINPS. Order quantities identified by the simulation provide higher expected profit than would have been obtainable without considering substitutability.

The purpose of this paper is to employ statistical methods such as cluster, discriminant, and other multivariate techniques to classify discrete unreliable transfer lines and to further evaluate how design and operational factors affect their performance. The empirical analysis to illustrate the statistical approach is based on a model and an experimental analysis derived by Mehrez and Patuwo

Analyses the economic efficiency of innovation under different market conditions within the context of a formal model that represents competitive innovation as the closed loop solution of a dynamic stochastic game in the spirit of Reinganum’s analysis. Goes beyond Reingnaum’s model in its explicit treatment of the product market (following Spence), which permits a consistent analysis of competitive innovation in the absence of patent protection. Measures efficiency by the expected net present value of total surplus, and studies the effect on it of competitive structure, elasticity of demand, fixed costs, spillovers and subsidies.

Many extensions to the newsboy problem have been solved in the literature. Two of those extensions solve a multi-product newsboy problem with a storage or budget constraint, and a single-product newsboy problem in which multiple discounts are used to sell excess inventory. In this paper, we formulate and solve a multi-product newsboy problem in which multiple discounts are used to sell excess inventory under a storage or budget constraint. We provide an efficient algorithm to solve the new problem and illustrate the results with numerical examples.

This paper presents a two-stage model for industrial robot operation and selection in a multi-product manufacturing system. The quality of the production process is related to robot repeatability and the output rate is related to robot speed. Robot repeatability deteriorates with increased speed. In the second stage, a one-period robot operational model maximises the expected profit of a risk-neutral decision maker, leading to optimal robot speed and production quantity decisions. Using the optimal speeds and quantities of the operational model and taking into account demand uncertainties, the first stage robot selection model is solved.

This paper describes the first dynamic formulation of the optimal strategy, from a patient perspective, for the search for additional information in the form of multiple opinions. The Mehrez and Gafni framework for choosing an optimal treatment from a patient perspective is employed to study the optimal search strategy for a representative individual who is an expected utility maximizer. Two different processes for gathering information regarding the probability of treatment success ( P ) are studied: In the first, physicians provide information in a yes or no (treatment) type of recommendation; in the second, physicians provide probabilistic information about the likelihood of success in the treatment. For the first system the optimal strategy is derived using an algorithm of an order O( n ) ( n is the number of opinions to be purchased from expert doctors) for a prior beta probability density function of P and a Bernoulli process of information gathering. EVSI (expected value of sample information) measures are incorporated into the Bayesian analysis to study this case and to obtain an upper bound on n ∗ (the optimal number of opinions to be purchased). The approach developed here is different than the well known one provided by Raiffa and Schlaifer in 1961 and it enables us to present additional results regarding the optimal search strategy and how to identify it. For example, if (i) the process of information gathering is generated by a Bernoulli process, (ii) the natural conjurate prior distribution is beta, and (iii) the utility functions are linear functions of the probability of treatment success, then for each n , there exists at most one value of x (the number of experts recommending treatment) for which a search should continue. At this value the incremental increase of the individual's lifetime expected utility stemming from the decision to seek additional expert's opinion is positive. This observation might explain a reality where the likelihood of seeking a large number of opinions is very low. Furthermore, sufficient conditions for not initiating the information search process and on the range of n ∗ are provided. In addition an analysis of the second process is provided and properties of the optimal strategy are studied. It is shown that the formulations of the two processes are equivalent and that they can be further analyzed through the two action space project selection problem studied by Mehrez and Stulman and others. A numerical comparison and a sensitivity analysis between the two processes for gathering information is presented. It is illustrated that for similar cost figures, the probability estimation procedure provides more information than the yes-no type recommendation to the patient.

In this paper, we consider the partial backorder policy used in conjunction with the traditional stochastic (r, Q) inventory system. The partial backorder control is modeled using a control variable, b, which limits the maximum number of backorders allowed to be accumulated in any given cycle. In traditional inventory control policies, the unsatisfied demands are either completely backordered or completely lost. The new control variable b provides an alternative to dealing with shortages, which is different from the two extremes policies: the pure backorder policy (which corresponds to b=∞) and the lost sales policy (b=0). We obtain the expected annual cost of the model and give a procedure to compute the optimal parameters, r∗, b∗ and Q∗. Numerical examples which demonstrate the advantage of implementing the partial backorder policy are presented and the percentage of cost savings depends on the fill rate.

This paper reports the modeling and solution of an industrial ocean-cargo shipping problem. The problem involves the delivery of bulk products from an overseas port to transshipment ports on the Atlantic Coast, and then over land to customers. The decisions made include the number and the size of ships to charter in each time period during the planning horizon, the number and location of transshipment ports to use, and transportation from ports to customers. The complexity of this problem is compounded by the cost structure, which includes fixed charges in both ship charters and port operations. Such a large scale, dynamic, and stochastic problem is reduced to a solvable stationary, deterministic, and cyclical model. The process of modeling the problem and the solution of the resultant mixed integer program are described in detail. Recommendations from this study have been implemented.

Recently Mehrez et al., Mehrez and Stern and Melkman et al. studied a vehicle fleet refueling problem that arises in military applications and is aimed at maximizing the operational range of the fleet. More specifically, they investigated the problem of maximizing the range of the last vehicle from a fleet of n vehicles by employing a pure refueling chain strategy. However, the problem is solved only for the vehicles with the same fuel consumption rate or the same fuel capacity. This paper is aimed at solving the problem of a more general case where vehicles may have different fuel consumption rates and different fuel capacities. Based on the analysis of fuel demand and supply curve, a recursive procedure which requires only O(n) calculations is developed to solve the linear programming problem of maximizing the operational range for a given refueling chain. A pairwise interchange rule is also provided to improve the maximum operation range by switching the adjacent vehicles in a refueling chain. Finally, a heuristic approach is suggested for finding an optimal refueling chain. Computer simulations show that for n = 6, by using the heuristic approach, the chance of finding an optimal solution is 72% and the average operational range is 97.8% of the optimal. Furthermore, employing an O(n) recursive formula, the solution of a multiple-transfer refueling chain is derived. An illustrative example is demonstrated for a military assessment of the trade-offs between the maximum operational range and the number of fuel transferring operations under the pure and the multiple-transfer chains.

This paper is concerned with an identification problem of a set of K out of N numbers. Two parties are involved in the identification process, the inquirer and the respondent. The set to be identified is referred to as the chosen subset and it is known only to the respondent. The identification process involves the inquirer who wishes to identify the chosen subset by asking a series of questions. At each question the inquirer selects a subset of K elements out of N, the inquired subset, and receives from the respondent partial information about the number of elements which belongs to both the inquired subset and the chosen subset. The questions continue until the chosen subset is identified. A possible application of the problem is related to learning games. Two types of algorithms, which are independent of each other and aimed at identifying the chosen subset, are presented in this paper. Type I algorithms (uncertainty reduction) select only subsets that have not yet been rejected as being the chosen subset. A linear integer programming model is employed to identify the set of subsets that have not yet been rejected—the uncertainty set. For Type I algorithms two different decision rules are considered to select a subset. Each decision rule selects the elements of the next inquired subset by employing a different probabilistic argument: 1) randomly select the next subset from among the subsets not yet ruled out, or 2) select as the next subset the one whose elements occur most frequently among the subsets not yet ruled out. Type II algorithms (static and dynamic processing) select a subset according to a different principle which allows the inquirer to select a subset that may not belong to the uncertainty set. This is accomplished by marginally interchanging the elements of any two successive subsets. Analytical solutions are provided for algorithms of Type II. Furthermore, a comparison between the two types of algorithms under different measures of performance, such as the expected number of questions (subsets) required to identify the chosen subset, is conducted through a simulation analysis. INFORMS Journal on Computing, ISSN 1091-9856, was published as ORSA Journal on Computing from 1989 to 1995 under ISSN 0899-1499.

Certain management decision problems are not often easily modeled using traditional decision-making frameworks. Complex management decision problems may possess characteristics that are difficult to describe fully with any single decision-making framework such as a decision tree or a network model. A Petri Net is a powerful and a highly versatile modeling tool that possesses meta-model capability to replace alternative decision-making frameworks and model complex management decision problems or situations. The purpose of this paper is two-fold: (1) to review the properties of Petri Net models in view of management perspectives; and (2) to illustrate their power to describe various decision-making frameworks including decision trees, network models, Markov decision problems and time-table problems. Thus, we illustrate how Petri Net models possess meta-model capability to replace alternative decision-making frameworks.

This work reports on a mathematical model dealing with the optimal development of marginal water sources in the Negev Desert, a region located in the south of Israel, inhabited by less than one-tenth of the whole population of Israel, and consisting of about half of its area. A decomposable mixed linear zero one integer programming problem is formulated and analyzed to integrate decision-making at the local and regional levels. At the regional level, the problem is one of allocating a limited supply of high quality water from a primary regional source among a number of local demand sites. Each demand site must then determine the optimal investment strategy for developing its local marginal water sources in order to make up the deficit in supply and meet net water quality and budgetary requirements at the site. The relationship between the local and regional levels of decision-making is studied, and a procedure to solve the combined problem is proposed. A small case study is presented to illustrate the model.

This paper presents an optimization model for maximizing the insurer’s underwriting profits in any particular line of insurance at a point in time. We propose the use of mathematical programming to maximize underwriting profits when the objective function is expressed as a nonlinear convex function subject to linear inequality constraints. The constraints are imposed on premiums, loss payments and resources allocated for underwriting and loss adjustment activities. Although basic global solution is not guaranteed, under a set of sufficient conditions global basic solution is identified. An alternative basic solution is obtained where maximization is done over an expression which contains both the expected value as well as the variance of profits. Several nonlinear algorithms are outlined for possible solutions.

This paper presents a case study where academic departments at Ben-Gurion University were evaluated via the Data Envelopment Analysis using the CCR model. Extensive post analyses were performed in several directions. First various sets of data were used to identify efficient and inefficient departments. New efficiency measures are suggested in relation to the reference set included in the analyses of academic departments. We measured the efficiency of departments to other departments within the same school. We applied cluster analyses to divide the departments into several sets; and the discriminant analysis to test the match of the efficiency/inefficiency division of the CCR ratio. We further tested organizational changes where an inefficient department was closed and joins other departments. Finally we compared the CCR model to the pure economic approach—the cost per student ratio.

We evaluate the performance of unreliable production lines with intermediate buffers facing a random unit demand which causes blocking and starvation to the production line. It is assumed that the stations in the line are inspected periodically and if a station fails to meet predetermined operational standards, it is shut down for that entire period. To increase output, finite size buffers are placed in between stations and after the last station. A Markov chain is used to model the system and an efficient algorithm to compute the marginal limiting distributions of the buffer contents is presented. These limiting probabilities are then used to study the behavior of the system via experimental analysis. This paper also demonstrates the determination of the design variables (capacities, station reliabilities) and operational variables (work-in-process, capacity utilization) that do not have a unique optimal pattern. This is done in contrast to optimization via Markov decision processes.

A mixed binary integer linear program is formulated to determine the economic development of marginal groundwater sources at local demand sites in an arid region. These marginal sources are required to augment the supply from an overloaded regional source. The model accounts for variable costs of supply, fixed investment costs, capacity constraints at the regional and local levels, and water quality requirements at the local sites. A Lagrangian relaxation reduces the model to a series of simple local problems, the solution of which provides an optimal sequence for developing the marginal groundwater sources while reducing the demands on the regional source. A heuristic and an exact procedure are also proposed to solve the problem for arbitrary levels of supply from the regional source. The exact procedure uses characteristics of the optimal solution to reduce the model to a series of knapsack-type problems. The theory is applied to a small case study taken from the Negev Desert in southern Israel.

A well-known problem of prediction in linear regression models is to find a confidence interval for the random value of the dependent variable when the values of the independent variables are given. Such a situation may arise in economic quality control models when the independent variables are costly inputs and the dependent variable is some measure of quality or production. In such a circumstance, an important control objective may be to find values for the inputs that will maximize the lower limit of the prediction confidence interval for a fixed budget, or alternatively, to minimize the cost of the inputs for a fixed lower limit of the confidence interval. In this paper, we shall show that global optima can be found using known algorithms. The special case of simple linear regression is discussed and an illustrative example is provided.

We obtain an optimal operation policy for a general n-server (channel, machine) stand-by system involving preventive maintenance and operating in real time. We show that such a policy is achieved by successive total use of each server capacity, and does not depend on the arrival pattern of processed data and task duration. We show also that this policy is optimal for any finite, initial period of time, and not only in the sense of long-run availability. The analysis is performed in two stages: a worst-case analysis is performed, and then, a general case is treated. Such a consecutive approach also seems to be useful in the analysis of more complicated systems. Rigorous mathematical proofs are provided. Actual and possible applications of results to military and production control systems are discussed.

An optimal operation policy for n -server stand-by systems operating in real time is obtained. It appears that, under certain conditions, such a policy is achieved by successive total use of each server capacity. The analysis is performed in two stages: (1) a non-stop operating system is treated, and then, (2) a general system, which operates only in response to orders. As demonstrated, such a consecutive approach seems to be highly promising for the analysis of complicated systems involving preventive maintenance and operating in real time.

A methodology for information system planning and selection among alternative solutions for computerization of an organization is presented. The methodology is composed of the following stages: 1. Definition of the organizational processes and objective. 2. Definition of the computer-related goals set for each organizational objective. 3. Definition of the alternative information system processes required for each of the computer-related goals. 4. Identification of alternative systems to satisfy those information system processe and selection of the preferred system. 5. Preparation of a request for proposal. 6. Selection of the preferred alternative

The coordination of logistics decisions such as maintenance, replacement, inventory and transportation in a single framework is a very difficult task. This paper is motivated by an observation about the decision making process of a multi-echelon military system. We formulate a minimization problem in order to optimize logistics decisions given that static equilibrium environmental and organizational conditions are set. These refer to a deterministic stationary demand, fixed number of machines, exponential probability distribution function of failure and repair, a periodic transportation tour and EOQ inventory policies as well as other characteristics. Furthermore, the problem is reformulated as an unconstrained single variable problem and a procedure to identify its local optimal solutions is suggested. Also, the results of an experimental study consisting of 10,000 runs carried out on an IBM PC computer is reported with analytic emphasis on the cost components under different assumptions regarding the data structure. Finally, conclusions are drawn with respect to policy implications of the model.

Much work has been directed to develop aggregate efficiency measures for firms or decision-making units (DMUs) in which we are able to observe only the outputs and inputs. Assuming that each DMU has the same type of observed outputs and inputs and using only this information, Farrell's technical efficiency and the CCR ratio can be used to assign an aggregate measure of efficiency to each DMU, which can then be used to compare the efficiency of the DMUs. This paper considers a subset of the general aggregate efficiency problem called the matched output/input case in which each output is matched to exactly one input, forming a subunit. Dividing the output by the input for each subunit within a DMU yields a subunit ratio that is the output per unit input. For a particular subunit, the subunit ratios for two DMUs can be compared directly. If all the subunit ratios of one DMU exceed the corresponding subunit ratios in another DMU, then we should reasonably expect that any aggregate efficiency measure has the efficiency of the first DMU greater than the efficiency of the other DMU. This requirement is defined as the Matched Output/Input Axiom, which is then shown to be violated for certain data sets satisfying Simpson's Paradox. Both Farrell's technical efficiency and the CCR ratio are then shown to violate the Matched Output/Input Axiom, which raises questions about the overall validity of both procedures. Resume. Les travaux visant l'elaboration de mesures globales du rendement des unites decisionnelles ou des entreprises, dans lesquelles il n'est possible d'observer que les extrants et les intrants, sont nombreux. En supposant que le meme type d'extrants et d'intrants est observe pour chaque unite decisionnelle et que cette information est la seule qui soit utilisee, le rendement technique de Farrell et le ratio CCR (Charnes, Cooper et Rhodes) peuvent etre utilises pour attribuer une mesure globale de rendement a chaque unite decisionnelle, mesure qui peut ensuite permettre de comparer le rendement des differentes unites. Les auteurs etudient un sous-ensemble du probleme general de rendement global, le cas de concordance extrant-intrant, dans lequel chaque extrant est associe a exactement un intrant, pour former un sous-ensemble. En divisant l'extrant par l'intrant pour chaque sous-ensemble d'une unite decisionnelle, on obtient, pour chacun d'eux, un ratio representant l'extrant par unite d'intrant. Pour un sous-ensemble particulier, les ratios de deux unites decisionnelles peuvent faire l'objet d'une comparaison directe. Si la totalite des ratios des sous-ensembles d'une unite decisionnelle excede la totalite des ratios des sous-ensembles correspondants d'une autre unite decisionnelle, on est en droit de s'attendre a ce que l'application d'une mesure globale du rendement, quelle qu'elle soit, indique que le rendement de la premiere unite est superieur a celui de la seconde. Les auteurs demontrent que ce principe, defini comme etant l'axiome de concordance extrant-intrant, est transgresse pour certains ensembles de donnees repondant aux criteres du paradoxe de Simpson. Ils demontrent egalement que le rendement technique de Farrell ainsi que le ratio CCR derogent a l'axiome de concordance extrant-intrant, ce qui les amene a s'interroger sur la validite globale des deux procedes.

The problem of job shop scheduling, requires satisfaction of diverse and conflicting constraints. A large body of scheduling algorithms exists. The search for an appropriate algorithm or model can be very frustrating as well as discouraging for the prospective user, due to the complexity and the costs involved. Expert help may not always be affordable or accessible at the right moment. Knowledgebased Model Management Systems (MMS) enable us to make the expertise more accessible and affordable. In this paper we develop Scheduling Assistant, an interactive prototype knowledgebased MMS written in VAX OPS5 that displays a methodology for knowledgebased job shop scheduling decision support. It makes the search for expert scheduling knowledge easier, efficient, and more accessible to the practitioner user.

This study employs a descriptive model to relate key issues in inventory management — cost and demand estimation, lead time and performance measures of inventory systems. The model assumes a single item, uncertain demand, deterministic review order point and discrete lead time. Sixteen forecasting methods were examined and compared on empirical ground with three data sets. Cluster analysis was first used to group forecasting methods along four defined dimensions. The naive forecasting method was found to be distinctly different from the rest. Furthermore, the naive method consistently provided less accurate forecasts across the three data sets used. Cluster analysis was again used to put the other fifteen forecasting methods into groups. Results indicated that under general conditions regarding the demand pattern none of the forecasting methods have demonstrated the expected advantage relative to other methods. Thus more specific knowledge is required to assign a forecasting method to specific data. Such a model was recently studied by Reinstein, Gibson and Zidman under the naive forecasting method which calls for inventory based on the demand from the prior period. The results also have valuable insights for inventory management.

A two-stage decision and control model for the operation of the power station in the Hazbani-Dan Water System (Israel) was developed and implemented. In this system, river water is stored and then either consumed by agricultural farms or released to a hydroelectric power station. Under management policy, agricultural consumption is taken as a constraint. The problem was solved via a two-stage procedure consisting of a long-term model (LTM) and a short-term model (STM). The LTM was presented by Rabinowitz et al. (1988). This paper reports on the development and analysis of the STM, in which operational flexibilities of the power station are taken into consideration, thereby improving the LTM solution. Our main contributions are aimed at: (1) describing the (nontraditional) systematic thinking process which transformed the mathematically expressed problem into an efficiently solvable model; (2) presenting the analysis of the model; and (3) reporting on a practical application of the model in a unique system.

In this paper we consider a type of bias which stems from the mathematical algorithm often used to determine an optimal match between two groups. We compare two different solution concepts for the matching assignment problem: the traditionally stable solution vs. a multiplicative utility approach that should avoid the bias. Simulation modeling leads us to conclude that: (1) With respect to all sizes compared in our experiment, applicants and employers groups were always treated far more equally under the multiplicative utility approach than the stable approach. (2) When using the stable algorithm, the size of the bias is affected by the size of the problem (i.e., the larger the problem size is the larger the performance discrepancy between two groups of participants). Further analyses of the sensitivity of the finding to different assumptions (e.g., correlation in the preference orderings of the two groups, or use of nonlinear conversion from ranks to utilities) also resulted in a superior performance of the multiplicative utility algorithm in terms of a more equitable outcome.

Most inventory models assume that the production lot size does not contain any defectives. Clearly, this assumption does not reflect the reality of most manufacturing systems. The number of defects in a production lot depends on the manufacturing technology implemented by the firm. In this paper we study the decision of a firm seeking to replace its existing manufacturing technology with an improved one. Furthermore, we study the effect the quality of the technology has on the optimal production lot size. In this paper, we formulate a two-stage stochastic programming problem. In the second stage, the optimal production lot size is derived for a given manufacturing technology and a given demand forecast. This derivation is provided under the framework of the one-period inventory problem while maintaining a minimal service level. In the first stage, and using the optimal production lot size of the second stage problem, an optimal manufacturing technology that minimizes the total expected cost to the firm is implemented. An algorithm to find an optimal solution is presented. A numerical example illustrates the results.

The Hazbani-Dan water system supplies gravitational water for both irrigation and hydroelectric energy production. An automatic control system for operating the energy plant while satisfying the irrigation needs was developed and applied. This was accomplished through an interdisciplinary research effort comprised of a feasibility study, formulation and analysis of a two level optimization model, development of forecasting models, and information system analysis. The value of such a control system goes beyond its direct economical benefits. It also provides tools for management decisions involving the evaluation of investments, and the determination of operating rules. This paper consolidates the narrow and technical scope of previous reports by presenting the overall research methodology, the relationships among the different parts, and a unique way of conducting applied research in an academic environment.

We address a practical problem faced by many firms. The problem is deciding on the production levels for a product that has a very short selling season. The firm has a full period to produce and meet a lumpy demand which occurs at the end of the period. The product is no longer demanded after the end of the period. A constant production rate which minimizes average unit cost may increase holding costs. Varying the production rate at discrete points in time may increase production costs but may also decrease holding costs. In addition, allowing changes in the production rate enables the incorporation of forecast revisions into the production plan. Therefore, the best production plan depends on the flexibility of the production system and on the holding cost. In this paper, we formulate and solve a model of this production planning problem. Two models are developed to deal with two types of the average unit cost function. Numerical examples are used to illustrate the results of the model.

This paper suggests a two-level game analysis of Israel’s strategy toward peace during the 1990s. The paper shows how various paradoxes in Israeli society create domestic obstacles and internal opposition that weaken Israel’s bargaining position toward neighboring countries. Treating domestic parameters in these countries as a given, we argue that Israeli leaders can hardly use this weakness to manipulate information in the bargaining process, because neighboring countries can observe Israel’s internal processes. Therefore, attempts by Israeli leaders to create the impression that they are willing to adopt a conflictual approach towards neighboring polities, especially the Palestinians, without actually creating the necessary internal conditions for such a policy, may finally lead to a sub-optimal equilibrium for Israel – in terms of territory and deterrent ability – since it will have to compromise under difficult conditions. Several practical implications as to the preferred bargaining process under these conditions follow.

Recently, Mehrez and Steinberg (1995) described and studied the matching identification problem (MIP). The MIP is a form of knowledge acquisition problem from the field of artificial intelligence. For instance, an expert system infers knowledge from a set of examples. But how do you most quickly acquire the examples that knowledge is inferred from? The MIP is a special case of this problem. Although an optimal algorithm was not found by Mehrez and Steinberg, they described two general types of heuristics. We describe in this paper an optimal algorithm for the case of K=2, and an improved heuristic for general K, which identifies a chosen subset with 6% fewer inquiries on average when N=15, K=3. The heuristic improves relative to the Type I heuristic as N increases, K held constant. The improved heuristic is concerned with the symbols yet unclassified as being in the chosen subset or not in the chosen subset. By inquiring subsets with all unclassified symbols, we most quickly "span" the set of unclassified numbers. Closed form equations are developed for the expected number of inquiries required and the variance of the number of inquiries required for the optimal algorithm. Computational studies are provided for Mehrez and Steinberg's Type I heuristics, the K=2 optimal algorithm, and the spanning heuristic.

Recent literature use logit analysis for point estimators of the probability of insolvency for each insurer based on financial statements data. These studies focus on selecting predictors, comparing classification techniques, and measuring classification accuracy. Our analysis differs from the previous literature by explicating and developing confidence intervals which provide measures of statistical reliability for the estimators and thus enhance the results of prior research. The main purpose of this article is to provide and illustrate a method of constructing confidence intervals for insolvency probabilities. We examine various measures of the confidence intervals, such as their minimum lengths and minimum upper bounds. Two examples show a substantial improvement (reduction) in the length and the minimum upper bound of the confidence intervals at the optimal level of the financial accounting variables. A third example depicts a confidence interval for the probability of failure for an insolvent insurer. Implications for researchers and users of financial accounting information are briefly discussed.

Recently a public committee has been appointed by the Israeli government. The committee was assigned to prepare regional plans for the northeastern Negev, in the southern part of Israel, in order to absorb some of the large waves of immigrants (about 400,000) expected by the year 2000. Among the strategic issues treated by the committee were the questions when, where and whether to build a new hospital in the area. The locational decision was studied by first identifying a finite candidate set of possible sites. At the final stage various quantitative locational models were employed. In the next stage a subjective multicriteria model was further used to derive locationally preferred solutions. This paper focuses on the first stage. Furthermore, the suggested solution approach is discussed in light of strategic issues.