Perishable Inventory Problem Research Articles

Surrogate-assisted evolutionary optimization for perishable inventory management in multi-echelon distribution systems

Simulation is widely used for analyzing supply chains with complex structures and stochastic nature. However, optimizing supply chain simulation models is usually computationally expensive. This study proposes a surrogate-assisted evolutionary optimization approach to optimize the inventory policies in multi-echelon distribution systems for perishable items under a limited number of evaluations. The random forest algorithm is used to build the surrogate model for a faster estimation of the performance of the inventory policies. A co-evolutionary differential evolution algorithm is proposed to simultaneously evolve the population through multiple searching strategies. The generated solutions are estimated by the low-cost surrogate model to select the promising solutions, which will be evaluated by the inventory model. Moreover, this study also integrates the surrogate model into two classic metaheuristic algorithms, particle swarm optimization and differential evolution. Also, a new performance indicator is proposed to examine the efficiency of the surrogate model in evolutionary computation. Two case studies are used to investigate the performance of the proposed algorithms. The experimental results show that both particle swarm optimization and differential evolution exhibit performance improvements exceeding 55% by using surrogate models under the limited number of function evaluations. Furthermore, the surrogate model reduces computational time for both algorithms by over 34% to achieve equivalent objective values. Finally, the proposed co-evolutionary differential evolution algorithm is compared with 12 algorithms, and the results show that the proposed algorithm consistently outperforms them. These findings confirm the usefulness of the surrogate model in evolutionary algorithms and the effectiveness of the proposed co-evolutionary strategy for solving perishable inventory problems.

Heuristics and sensitivity analyses to guide replenishment decisions for red blood cells with random transfer

While the transfer of blood is common in practice, research addressing replenishment strategies in the context of random or unscheduled transfers is scarce and the few papers that consider ordering policies with multiple independent sources of supply offer mathematically complex solution approaches that are difficult to implement in practice. There are no simple heuristic approaches to guide ordering decisions in these situations. In this paper, we introduce two near-optimal heuristics to guide replenishment decisions for the fixed lifetime perishable inventory problem with random transfers. Much like the traditional base stock policy, the heuristics are straightforward to calculate and implement. The first heuristic (H1) employs a base stock policy using the convolution of demand and transfer distributions. The second heuristic (H2) employs a transfer adjusted base stock policy. We validate our heuristics using inventory-related cost components (i.e., order, holding, wastage, and underage costs) and demand and transfer distributions based on data provided by a large university medical center. Via simulation, we evaluate the performance of the heuristics by comparing the total costs obtained using each heuristic to the total costs obtained using the optimal ordering policy. Both heuristics perform quite well. We conduct a variety of sensitivity analyses considering the effects of different (1) levels of holding and stockout costs, as these costs are difficult to quantify and can vary dramatically, and (2) percentile transfer adjustments in H2. We observe that the H1 policy consistently results in slightly larger orders than the optimal policy, which leads to a slightly higher inventory, holding costs and wastage rates, yet slightly lower stockout costs. The H2 policy, on the other hand, consistently results in slightly smaller orders than the optimal policy, which leads to a slightly smaller inventory, lower holding costs and wastage rates, yet slightly higher stockout costs. To guide decision making, we discuss a number of management challenges that we encountered as part of this project and offer a five-step process to help decision makers with replenishment decisions.

Resource optimization for cancer pathways with aggregate diagnostic demand: a perishable inventory approach.

This work proposes a novel framework for planning the capacity of diagnostic tests in cancer pathways that considers the aggregate demand of referrals from multiple cancer specialties (sites). The framework includes an analytic tool that recursively assesses the overall daily demand for each diagnostic test and considers general distributions for both the incoming cancer referrals and the number of required specific tests for any given patient. By disaggregating the problem with respect to each diagnostic test, we are able to model the system as a perishable inventory problem that can be solved by means of generalized G/D/C queuing models, where the capacity [Formula: see text] is allowed to vary and can be seen as a random variable that is adjusted according to prescribed performance measures. The approach aims to provide public health and cancer services with recommendations to align capacity and demand for cancer diagnostic tests effectively and efficiently. Our case study illustrates the applicability of our methods on lung cancer referrals from UK's National Health Service.

Managing Perishable Inventory Systems as Non-Perishable Ones

Perishable inventory problems have a long history and involve two fundamental decisions, how much to order and how much old inventory to clear before expiration, that are known to be difficult to optimize due to the curse of dimensionality. Most early work ignores the clearance decision and focuses solely on the ordering decision until recently where heuristic clearance policies have been developed. In this paper, we approach the problem from a different angle by exploring its asymptotic behavior, i.e., perishability can be ignored in many cases and hence clearance of inventory is not necessary except at the beginning of the planning horizon when a system is large enough. Inspired by such asymptotic behavior, we examine simple policies that ignore clearance under minor conditions and establish theoretical bounds for them. The bounds not only vanish asymptotically, but also indicate a system size required to guarantee any given optimality gap. Numerical studies suggest that such policies can work very well for systems with reasonable sizes and practical management of complex perishable inventory systems is not so much harder than that of non-perishable ones.

A non-parametric adaptive algorithm for the censored newsvendor problem

This article studies the problem of determining stocking quantities in a periodic-review inventory model when the demand distribution is unknown. Moreover, lost sales are unobservable in the system and hence inventory decisions are to be made solely based on sales data. Both the non-perishable and perishable inventory problems are addressed. Using an online convex optimization procedure, a non-parametric adaptive algorithm that produces inventory policy in each period that depends on the entire history of stocking decisions and sales observations. With the help of a convex quadratic underestimator of the cost function, it is established that the T-period average expected cost of the inventory policy converges to the optimal newsvendor cost at the rate of O(log T/T) for demands whose expected cost functions satisfy an α-exp-concavity property. It is shown that, when the demand distribution is continuous, this property holds the probability density function over the decision set is bounded away from zero. For other continuous distributions, a “shifted” version of the density function is constructed to show an ε-consistency property of the algorithm so that the gap between the T-period average expected cost of the proposed policy and the optimal newsvendor cost is of the order O(log T/T) + ε (for a given small ε > 0). Simulation results show that the proposed algorithm performs consistently better than two existing algorithms that are closely related to the proposed algorithms.

Patient-based pharmaceutical inventory management: a two-stage inventory and production model for perishable products with Markovian demand

Drug shortages have increased over the past decade, tripling since 2006. Pharmacy material managers are challenged with developing inventory policies given changing demand, limited suppliers, and regulations affecting supply. Pharmaceutical inventory management and patient care are inextricably linked; suboptimal control impacts both patient treatment and the cost of care. We study a perishable inventory problem motivated by challenges in pharmaceutical management. Inpatient hospital pharmacies stock medications in two stages, raw material and finished good (e.g. intravenous). While both stages of material are perishable, the finished form is highly perishable. Pharmacy demand depends on the population and patient conditions. We use a stochastic ‘demand state’ as a surrogate for patient condition and develop a Markov decision process to determine optimal, state-dependent two-stage inventory and production policies. We define two ordering and production scenarios, prove the existence of optimal solutions for both scenarios, and apply this framework to the management of Meropenem.

Weak aggregating algorithm for the distribution-free perishable inventory problem

We formulate the multiperiod, distribution-free perishable inventory problem as a problem of prediction with expert advice and apply an online learning method (the Weak Aggregating Algorithm) to solve it. We show that the asymptotic average performance of this method is as good as that of any time-dependent stocking rule in a given parametric class.

Determination Of Outdate And Shortage Quantities In The Inventory Problem With Fixed Lifetime

The fixed lifetime perishable inventory system is examined and an estimator for the probability that an item will be sold in a period is obtained. The estimator for the probability that an item will be sold is given by \\vartheta / S where \\vartheta is the average demand per period and S is the inventory level at the start of a period. The probability is used to derive outdate and shortage quantities. Furthermore, shortage-outdate operating curve is obtained. The probability of running out of stock is obtained from the demand distribution. The operating characteristics obtained in this paper are very important because, for practical problems, available mathematically optimal solutions to the fixed lifetime perishable inventory problem cannot be realized due to their computational complexity. Guidelines for managing a real fixed lifetime perishable inventory system are given. A computer program has also been developed to compute the operating characteristics.

PERISHABLE INVENTORY PROBLEM WITH TWO TYPES OF CUSTOMERS AND DIFFERENT SELLING PRICES

This paper discusses an inventory control model for a single perishable product with two types of customers and different selling prices. This model is a one period horizon model and a generalization of Nahmias model [2] and Nose, et al. model [6] . Especially the model takes sensitive customers to freshness of commodities into consideration and treats two types of customers. Moreover, in the model there exist different, selling prices according to the remaining lifetime of commodity to be sold. In the situation, an optimal ordering policy to maximize the expected profit is derived.

An Application of the Theory of Weak Convergence to the Dynamic Perishable Inventory Problem with Discrete Demand

This paper develops a technique for analyzing dynamic inventory problems with discrete demand using the theory of weak convergence. We treat the dynamic fixed life perishable inventory problem with discrete demand. The optimal structure has been established assuming continuous demand. We construct an infinite sequence of continuous demand distributions that converge weakly to a discrete distribution, and show that the key results carry over in the limit.

An Inventory Model with Product Obsolescence and Its Implications for High Technology Industry

The advent of micro-computers and other high technology products, such as visual inspection systems and robots, with their supporting hardware/software components has been paralleled with an appreciable increase in the number of distribution outlets. Each commercial outlet has to carry enough inventory to provide proper service to its customers. However, due to the fast change in the technology involved, numerous inventory items become obsolete at a rapid rate. The situation can be treated as a perishable inventory problem. This paper develops that analogy, and presents the expressions for the: Required stock level, Optimal order quantity, and Total cost of the inventory policy. It is shown that as obsolescence rate increases, the Kanban (Just-in-Time) concept of purchasing, lot sizing, and inventories should be considered by the hierarchical organizations or the distribution network.

ON THE INVENTORY OF HIGH TECHNOLOGY PRODUCTS UNDER CONDITIONS OF DECAY

The advent of the micro-computers and other high technology products with its supporting hardware/software components has been paralleled with a significant increase in the number of distribution outlets. Each commercial outlet has to carry a certain amount of inventory so as to provide a proper service to its customers. However, due to the fast change in the technology involved, numerous inventory items become obsolete at an accelerating rate. The same case applies to manufacturers of high tech equipment such as programmable computers, visual inspection systems, robots, etc. As long as the “Kanban“ (Just-in-Time) inventory system has not been adopted, the organization is faced with this problem. The problem can be treated as a perishable inventory problem. In this paper the analogy is developed and expressions for the required stock level are presented.

Optimal Issuing Policies for Perishable Inventory

Optimal issuing policies for some particular classes of perishable inventory problems are derived under several possible objective functions. The inventory considered is one in which stock is grouped into categories according to shelf age. Demand occurs for each of the categories, and may be satisfied by inventory units from that category or from any “younger” category. It is shown for most of the objective functions considered that the optimal policy is to issue the oldest unit which will satisfy the demand. A prime example of the classes of inventory problems considered in this paper is the issuance of whole blood from a hospital or central blood bank.