Search

Urban logistic applications that involve pickup and distribution of goods require making routing and allocation decisions with respect to a set of sites. In cases where the supply quantities and the time in which they become available are unknown in advance, these decisions must be determined in real time based on information that arrives gradually. Furthermore, in many applications that satisfy the described setting, fair allocation is desired in addition to system effectiveness. In this paper, we consider the problem of determining a vehicle route that visits two types of sites in any order: pickup points (PPs), from which the vehicle collects supplies, and demand points (DPs), to which these supplies are delivered. The supply quantities offered by each PP are uncertain, and the information on their value arrives gradually over time. We model this problem as a stochastic dynamic routing and resource allocation problem, with the aim of delivering as many goods as possible while obtaining equitable allocations to DPs. We present a Markov decision process formulation for the problem; however, it suffers from the curse of dimensionality. Therefore, we develop a heuristic framework that presents a novel combination of operations research and machine learning and is applicable for many dynamic stochastic combinatorial optimization problems. Specifically, we use a large neighborhood search (LNS) to explore possible decisions combined with a neural network (NN) model that approximates the future value given any state and action. We present a new reinforcement learning method to train the NN when the decision space is too large to enumerate. A numerical experiment with 38 to 180 site instances, based on data from the Berlin Foodbank and randomly generated data sets, confirms that the heuristic obtains solutions that are on average approximately 28.2%, 41.6%, and 57.9% better than three benchmark solutions. Funding: This research was partially supported by the Israel Science Foundation [Grant 463/15], by the Shlomo Shmeltzer Institute for Smart Transportation at Tel Aviv University, by the Israeli Smart Transportation Research Center (ISTRC), and by the Council for Higher Education in Israel (VATAT). Supplemental Material: The online appendix is available at https://doi.org/10.1287/trsc.2023.0228. [ABSTRACT FROM AUTHOR]

Abstract. Bike sharing systems (BSSs) allow customers to rent bicycles at automatic rental stations distributed throughout a city, use them for a short period of time, and return them to [...]

We focus on a multi-period logistic setting, in which a given fleet of vehicles is used to collect products from suppliers, transfer them to a depot where they undergo certain time-consuming processing procedures, only at the end of which can they be distributed to customers. The problem is to schedule visits to the suppliers, as well as vehicle routes that distribute the processed products to the customers, such that each vehicle can perform at most one of these activity types every day. The processed inventory at the depot creates an additional dependence between these activities. Non-profit gleaning operations performed by food banks provide a real-life motivation for the analysis of this setting. It is also applicable for the collection and distribution of blood donations. In the solution method we propose, the problem is decomposed into its collection and distribution aspects, both of which constitute non-trivial sub-problems that have not been previously studied. We show how to tackle each of them individually, while considering information obtained from their counterpart, with the inventory storage as their linkage. We further present a rolling horizon framework for the problem. We demonstrate the implementation of this approach based on the activity of an Israeli food bank that manages gleaning operations, and test it with real-life data, as well as with randomly generated instances. The numerical experiments establish the advantage of the proposed method, compared to naïve methods currently being used in practice. [ABSTRACT FROM AUTHOR]

We consider a supply chain, which consists of several retailers and one supplier. The retailers, who possibly differ in their cost and demand parameters, may be coordinated through replenishment strategies and transshipments, that is, movement of a product among the locations at the same echelon level. We prove that in order to minimize the expected long-run average cost for this system, an optimal replenishment policy is for each retailer to follow an order-up-to S policy. Furthermore, we demonstrate how the values of the order-up-to quantities can be calculated using a sample-path-based optimization procedure. Given an order-up-to S policy, we show how to determine an optimal transshipment policy, using a linear programming/network flow framework. Such a combined numerical approach allows us to study complex and large systems., 1. Introduction Physical pooling of inventories (Eppen, 1979) has been widely used in practice to reduce costs and improve customer service. For example, Xerox has consolidated all of its country-based [...]

We consider a capacitated production-inventory problem in both discrete- and continuous-time, stationary settings. In the discrete-time setting we analyze the infinite horizon capacitated dynamic lot-sizing problem, find the optimal solution and characterize its properties. For the continuous-time setting we formulate a new problem, which we claim to be an appropriate counterpart of the above discrete-time problem. No other counterpart model was found in the literature, including the vast literature on optimal control, which presumably deals with similar problems but in a continuous-time framework. The new problem formulation is the basis for a new class of models, which forms an alternative way of analyzing certain dynamic lot-sizing problems. This new alternative could sometimes be simpler than the analysis in the discrete case., 1. Introduction When formulating an optimization problem, a modeler may choose either a discrete or continuous framework. In some cases, the derivation of the results may be simplified under one [...]

The problem addressed in this paper is the tool switching problem for an automated manufacturing environment, when each tool may occupy more than one slot of the tool magazine. A machine processes parts automatically by using a limited capacity tool magazine. Providing a tool that is needed for a certain processing operation is not in the magazine, a tool switch must occur before the job can be processed, a time/cost consuming operation. To solve this problem, one has to decide three types of decisions, namely, how to select the jobs' sequence (machine loading), which tools to switch before each processing operation (tool loading) and where to locate each tool in the magazine (slot loading). We present an integer programming formulation for the problem and suggest a heuristic procedure to obtain a solution. Our heuristic is partly a generalization of previously suggested approaches to the first two decision types, but it is mainly oriented towards answering the third decision type. The unified problem has not been addressed previously in the literature. We present a numerical study that demonstrates the efficiency of our procedure., 1. Introduction and literature review The concept of flexibility has become a key consideration in the design, operation and management of manufacturing systems. In a survey by Sethi and Sethi [...]

We consider centrally controlled multi-location systems, which coordinate their replenishment strategies through the use of transshipments. In a dynamic deterministic demand environment the problem is characterized by several locations, each of which has known demand for a single product for each period in a given finite horizon. We consider replenishment, transshipment and inventory holding costs at each location, where the first two have location-dependent fixed, as well as linear components, and the third is linear and identical to all locations. We prove that the resulting dynamic transshipment problem is NP-hard, identify a special structure which is satisfied by an optimal solution and develop, based on this structure, an exponential time algorithm to solve the problem optimally. In addition, we develop a heuristic algorithm, based on partitioning the time horizon, which is capable of solving larger instances than the optimal solution. Our computational tests demonstrate that the heuristic performs extr emely well., 1. Introduction and literature review Centrally controlled multi-location systems are capable of increasing their profitability through coordination. In recent years, organizations have started using novel logistic strategies such as internal [...]

Early diagnosis and treatment of newborns with Human Immunodeficiency Virus (HIV) can substantially reduce mortality rates. Polymerase chain reduction technology is desirable for diagnosing HIV-exposed infants and for monitoring the disease progression in older patients. In low- and middle-income countries (LMICs), processing both types of tests requires the use of scarce resources. In this article, we present a supply chain network model for referring/assigning HIV test samples from clinics to labs. These assignments aim to minimize the expected infant mortality from AIDS due to delays in the return of test results. Using queuing theory, we present an analytical framework to evaluate the distribution of the sample waiting times at the testing labs and incorporate it into a mathematical model. The suggested framework takes into consideration the non-stationarity in the availability of reagents and technical staff. Hence, our model provides a method to find an assignment strategy that involves an indirect prioritization of samples that are more likely than others to be positive. We also develop a heuristic to simplify the implementation of an assignment strategy and provide general managerial insights for operating sample referral networks in LMICs with limited resources. Using a case study from Tanzania, we show that the potential improvement is substantial, especially when some labs are utilized almost to their full capacity. Our results apply to other settings in which expensive equipment with volatile availability is used to perform crucial operations, for example, the recent COVID-19 pandemic. [ABSTRACT FROM AUTHOR]

We propose a new approach for the facility process-layout design problem, and introduce new mixed-integer linear programming (MILP) formulations for solving the problem. In this approach, we consider simultaneously, in a process-layout setting, the shape and location of the departments within the facility as well as the internal arrangement of the machines within the departments. Two models are suggested, the first assumes a rectangular shape of the departments and the second allows non-rectangular departments defined by an L/T shape. For the latter model, new constraints are developed to assure a correct design of the L/T shapes and to avoid irregular department shapes, and cuts are added to shorten the solution time. Finally, we conduct an extensive numerical study, in which we show the capabilities of both formulations in solving problems of medium size, and the superiority of the L/T department shape solutions over the rectangular department solutions. [ABSTRACT FROM PUBLISHER]

The general dynamic lot size model, also referred to as the Wagner-Whitin model, is examined. A simple algorithm which solves the general model in 0 (n log n) time and 0 (n) space is described. A linear algorithm is presented for two special cases: models without speculative carrying stock motives and models with setup costs that are nondecreasing. Conditions for monotone optimal policies are also derived and then related to known results.

In bike-sharing systems, at any given moment, a certain share of the bicycle fleet is unusable. This phenomenon may significantly affect the quality of service provided to the users. However, to date this matter has not received any attention in the literature. In this article, the users' quality of service is modeled in terms of their satisfaction from the system. We measure user dissatisfaction using a weighted sum of the expected shortages of bicycles and lockers at a single station. The shortages are evaluated as a function of the initial inventory of usable and unusable bicycles at the station.We analyze the convexity of the resulting bivariate function and propose an accuratemethod for fitting a convex polyhedral function to it. The fitted polyhedral function can later be used in linear optimization models for operational and strategic decision making in bike-sharing systems. Our numerical results demonstrate the significant effect of the presence of unusable bicycles on the level of user dissatisfaction. This emphasizes the need to have accurate real-time information regarding bicycle usability. [ABSTRACT FROM AUTHOR]