An examination of the size of orders from customers,their characterisation and the implications for inventory control of slow moving items

This paper examines half a million observations of the size of orders from customers at an electrical wholesaler. It notes: the distribution of the size of customer orders for a single item (stock keeping unit or SKU) is very skewed and resembles a geometric distribution; while the average size of an order is different for different items, for one SKU the mean order size is effectively the same at different branches even when the branches have very different demand rates; across a range of SKUs there is a strong relationship linking the mean and the variance of order size. The general results above are shown to apply to even the slowest movers. This extension is important because for items with intermittent demand the size of customer orders is required to produce an unbiased estimate of demand. Also a knowledge of the distribution of demand is important for setting maximum and minimum stock levels and the scheme employed is described.     

Optimal inventory policy with supply uncertainty and demand cancellation

We consider a periodic review model where the firm manages its inventory under supply uncertainty and demand cancellation. We show that because of supply uncertainty, the optimal inventory policy has the structure of re-order point type. That is, we order if the initial inventory falls below this re-order point, otherwise we do not order. This is in contrast to the work of Yuan and Cheung (2003) who prove the optimality of an order up to policy in the absence of supply uncertainty. We also investigate the impact of supply uncertainty and demand cancellation on the performance of the supply chain. Using our model, we are able to quantify the importance of reducing the variance of either the distribution of yield or the distribution of demand cancellation. The single, multiple periods and the infinite horizon models are studied.     

Inventory sharing via circular unidirectional chaining

This research studies the performance of circular unidirectional chaining – a “lean” configuration of lateral inventory sharing among retailers or warehouses – and compares its performance to that of no pooling and complete pooling in terms of expected costs and optimal order quantities. Each retailer faces uncertain demand, and we wish to minimize procurement, shortage and transshipment costs. In a circular unidirectional chain all retailers are connected in a closed loop, so that each retailer can cooperate with exactly two others as follows: receive units (if needed?available) from the left “neighbor” and send units (if needed?available) to the right, and a retailer who receives units from one neighbor is not allowed to send any units to its other neighbor. If the chain consists of at least three nodes and demands across nodes are i.i.d., its performance turns out to be independent of the number of nodes. The optimal stocking is therefore solved analytically. Analytical comparative statics with respect to cost parameters and demand distributions are provided. We also examine thoroughly the cases of uniform demand distribution (analytically) and normal demand distribution (numerically). In the uniform case with free transshipment, a unidirectional chain can save up to 1/3 of the expected cost of separate newsvendors caused by uncertainty. For three nodes, the advantage of complete pooling over unidirectional chaining does not exceed 19%.     

Dynamic fleet scheduling with uncertain demand and customer flexibility

We develop a dynamic fleet scheduling model that demonstrates how a carrier can improve fleet utilization. The fleet scheduling model presented by Lee et?al. (Eur J Oper Res 218(1):261–269, 2012) minimizes (1) a carrier’s fleet size and (2) the penalty associated with the alternative delivery times selected. The model is static since requests are collected over time and processed together. In this paper we present a stochastic, dynamic version of the fleet reduction model. As demand is revealed throughout an order horizon, decisions are made in stages by sampling anticipated demand to avoid recourse penalties in later stages. Based on computational experiments we find the following:

1. Modeling stochasticity improves the quality of solutions relative to the analogous model that does not include stochasticity. Counter-intuitively, an order lead-time distribution in which most loads are requested early can negatively impact optimal solution costs.
2. The stochastic model produces good results without requiring prohibitively large numbers of demand scenarios.
3. Consignees that place orders early in the order horizon are more often assigned their requested delivery times than those who place orders late.

     

离散模糊需求报童问题的可能性模型研究

基于可能性分布函数质心特征值,本文建立了确定离散模糊需求报童问题订货量的利润模型,并分析了成本模型和利润模型的关系。研究结果表明:1)基于可能性分布函数质心的模糊可能性成本和利润模型确定的订货量不一致;2)对应不同订货量,模糊可能性成本与利润之和不为固定常数。数值计算表明:该方法不可取。     

Unreliable newsboy problem with a forecast update

In this paper, we derive an optimal ordering policy for an unreliable newsboy who can place two sequential orders before the start of a single selling season by using a demand forecast update. Supply yield is modeled using a uniform distribution considering both the minimum order guarantee and the maximum yield. Our results indicate that a firm should focus on increasing the minimum order guarantee from a first stage supplier to reduce its total supply chain cost.     

Reducing lost-sales rate on the stochastic inventory model with defective goods for the mixtures of distributions

In this paper, we assume that the demands of different customers are not identical in the lead time. Thus, we investigate a continuous review inventory model involving controllable lead time and a random number of defective goods in buyer’s arriving order lot with partial lost sales for the mixtures of distributions of the lead time demand to accommodate more practical features of the real inventory systems. Moreover, we analyze the effects of increasing investment to reduce the lost sales rate when the order quantity, reorder point, lost sales rate and lead time are treated as decision variables. In our studies, we first assume that the lead time demand follows the mixture of normal distributions, and then relax the assumption about the form of the mixture of distribution functions of the lead time demand and apply the minimax distribution free procedure to solve the problem. By analyzing the total expected cost function, we develop an algorithm to obtain the optimal ordering policy and the optimal investment strategy for each case. Finally, we provide numerical examples to illustrate the results.     

Optimal ordering policies in inventory systems with random demand and random deal offerings

In this paper, we determine the optimal order policies for a firm facing random demand and random deal offerings. In a periodic review setting, a firm may first place an order at the regular price. Later in the period, if a price promotion is offered by the supplier (with a certain probability), the firm may decide to place another order. We consider two models in the paper. In the first model, the firm does not share the cost savings (due to the promotion offered by the supplier) with its own customers, i.e. its demand distribution remains fixed. In the second model, the cost savings are shared with the final customers. As a result, the demand distribution shifts to the right. For both the models, in a dynamic finite-horizon problem, the order policy structure is divided into three regions and is as follows. If the initial inventory level for the firm exceeds a certain threshold level, it is optimal not to order anything. If it is in the medium range, it is optimal to wait for the promotion and order only if it is offered. The order quantity when the promotion is offered has an ‘order up to’ policy structure. Finally, if the inventory level is below another threshold, it is optimal to place an order at the regular price, and to place a second order if the promotion is offered. The low initial inventory level makes it risky to just wait for the promotion to be offered. The sum of the order quantities in this case has an ‘order up to’ structure. Finally, we model the supplier's problem as a Stackelberg game and discuss the motivation for the supplier to offer a promotion for the case of uniform demand distribution for the firm. In the first model (when the firm does not share the cost savings with its customers), we show that it is rarely optimal for the supplier to offer a promotion. In the second model, the supplier may offer a promotion depending on the price elasticity of the product.     

Alternative formulations of a combined trip generation,trip distribution,modal split,and trip assignment model

The traditional four-step model has been widely used in travel demand forecasting by considering trip generation, trip distribution, modal split and traffic assignment sequentially in a fixed order. However, this sequential approach suffers from the inconsistency among the level-of-service and flow values in each step of the procedure. In the last two decades, this problem has been addressed by many researchers who have sought to develop combined (or integrated) models that can consider travelers’ choice on different stages simultaneously and give consistent results. In this paper, alternative formulations, including mathematical programming (MP) formulation and variational inequality (VI) formulations, are provided for a combined travel demand model that integrates trip generation, trip distribution, modal split, and traffic assignment using the random utility theory framework. Thus, the proposed alternative formulations not only allow a systematic and consistent treatment of travel choice over different dimensions but also have behavioral richness. Qualitative properties of the formulations are also given to ensure the existence and uniqueness of the solution. Particularly, the model is analyzed for a special but useful case where the probabilistic travel choices are assumed to be a hierarchical logit model. Furthermore, a self-adaptive Goldstein–Levitin–Polyak (GLP) projection algorithm is adopted for solving this special case.     

Inventory model involving controllable backorder rate and variable lead time demand with the mixtures of distribution

In the model of Ouyang and Chuang [Comput. Ind. Eng. 40 (2001) 339], they assume that the backorder rate is dependent on the length of lead time through the amount of shortages and let the backorder rate is a control variable. But, since they only assumed a single distribution for the lead time demand, when the demand of the different customers are not identical in the lead time, then we cannot use a single distribution (such as [Comput. Ind. Eng. 40 (2001) 339]) to describe the demand of the lead time. Hence, in our studies, we first assume that the lead time demand follows a mixtures of normal distribution, and then we relax the assumption about the form of the mixtures of distribution functions of the lead time demand and apply the minimax distribution free procedure to solve the problem. We develop an algorithm procedure, respectively, to find the optimal order quantity and the optimal lead time. Furthermore, two numerical examples are also given to illustrate the results.     

The setting of profit targets for target oriented divisions

Setting profit targets and striving to achieve them is fundamental to business survival and success. However, there has been little research on modeling profit-target setting. In this paper, we study analytic target setting under a common business scenario where a firm is in control of multiple divisions. Both the firm and the divisions maximize the profit probability, i.e., the probability of achieving some given profit target. The firm sets a profit target for each division which then acts as a price-setting newsvendor. We first obtain the optimal order quantity, the optimal retail price, and the maximal profit probability of a single division given its assigned target. We then derive the firm’s profit probability and focus on two specific cases to gain more managerial insights. In the first case of fair target setting, we show that when each division’s demand distribution has an increasing failure rate, if a division has a relatively high (low) production cost, its assigned profit target decreases (increases) with regard to its price elasticity. In the second case, if the firm is in control of two identical divisions, each division’s optimal profit target is just half of the firm’s profit target when the price elasticity is two or more, regardless of production cost and demand distribution.     

On the value of customer information for an independent supplier in a continuous review inventory system

We consider the inventory control problem of an independent supplier in a continuous review system. The supplier faces demand from a single customer who in turn faces Poisson demand and follows a continuous review (R, Q) policy. If no information about the inventory levels at the customer is available, reviews and ordering are usually carried out by the supplier only at points in time when a customer demand occurs. It is common to apply an installation stock reorder point policy. However, as the demand faced by the supplier is not Markovian, this policy can be improved by allowing placement of orders at any point in time. We develop a time delay policy for the supplier, wherein the supplier waits until time t after occurrence of the customer demand to place his next order. If the next customer demand occurs before this time delay, then the supplier places an order immediately. We develop an algorithm to determine the optimal time delay policy. We then evaluate the value of information about the customer’s inventory level. Our numerical study shows that if the supplier were to use the optimal time delay policy instead of the installation stock policy then the value of the customer’s inventory information is not very significant.     

Estimating retail demand with Poisson mixtures and out‐of‐sample likelihood

Estimation of retail demand is critical to decisions about procuring, shipping, and shelving. The idea of Poisson demand process is central to retail inventory management and numerous studies suggest that negative binomial (NB) distribution characterize retail demand well. In this study, we reassess the adequacy of estimating retail demand with the NB distribution. We propose two Poisson mixtures—the Poisson–Tweedie family (PTF) and the Conway–Maxwell–Poisson distribution—as generic alternatives to the NB distribution. On the basis of the principle of likelihood and information theory, we adopt out‐of‐sample likelihood as a metric for model selection. We test the procedure on consumer demand for 580 stock‐keeping unit store sales datasets. Overall the PTF and the Conway–Maxwell–Poisson distribution outperform the NB distribution for 70% of the tested samples. As a general case of the NB model, the PTF has particularly strong performance for datasets with relatively small means and high dispersion. Our finding carries useful implications for researchers and practitioners who seek for flexible alternatives to the oft‐used NB distribution in characterizing retail demand. Copyright © 2013 John Wiley & Sons, Ltd.     

On the exact calculation of the fill rate in a periodic review inventory policy under discrete demand patterns

The primary goal of this paper is the development of a generalized method to compute the fill rate for any discrete demand distribution in a periodic review policy. The fill rate is defined as the fraction of demand that is satisfied directly from shelf. In the majority of related work, this service metric is computed by using what is known as the traditional approximation, which calculates the fill rate as the complement of the quotient between the expected unfulfilled demand and the expected demand per replenishment cycle, instead of focusing on the expected fraction of fulfilled demand. This paper shows the systematic underestimation of the fill rate when the traditional approximation is used, and revises both the foundations of the traditional approach and the definition of fill rate itself. As a result, this paper presents the following main contributions: (i) a new exact procedure to compute the traditional approximation for any discrete demand distribution; (ii) a more suitable definition of the fill rate in order to ignore those cycles without demand; and (iii) a new standard procedure to compute the fill rate that outperforms previous approaches, especially when the probability of zero demand is substantial. This paper focuses on the traditional periodic review, order up to level system under any uncorrelated, discrete and stationary demand pattern for the lost sales scenario.     

Contracting with asymmetric demand information in supply chains

We solve a buyback contract design problem for a supplier who is working with a retailer who possesses private information about the demand distribution. We model the retailer’s private information as a space of either discrete or continuous demand states so that only the retailer knows its demand state and the demand for the product is stochastically increasing in the state. We focus on contracts that are viable in practice, where the buyback price being strictly less than the wholesale price, which is itself strictly less than the retail price. We derive the optimal (for the supplier) buyback contract that allows for arbitrary allocation of profits to the retailer (subject to the retailer’s reservation profit requirements) and show that in the limit this contract leads to the first-best solution with the supplier keeping the entire channel’s profit (after the retailer’s reservation profit).     

The effect of risk aversion on a supply chain with postponed pricing

We consider a supply chain consisting of a supplier and a risk-averse retailer operating under endogenous demand in retail pricing. The demand potential is uncertain and is revealed at the beginning of the selling season when it is too late to order products. The product price, on the other hand, is not determined in advance and can be postponed until the demand is revealed. The goal is to study the effect of risk-aversion and postponed pricing on both the retailer’s decisions and the overall supply chain. We find that the risk-averse retailer does not necessarily order less than the risk-neutral one and may introduce a bias by choosing a specific demand distribution. We contrast two specific choices. One is symmetric (balanced) with respect to the mean demand potential. The other is skewed (pessimistic) with most observations expected below the mean demand potential. Our numerical results show that the binding downside risk constraint deteriorates the supply chain performance when the forecast is balanced and improves it when the forecast is pessimistic.     

Dynamic Programming and Inventory Control

An expository account is given of the Dynamic Programming approach to inventory analysis in terms of the problem of the optimal storage policy for repair parts. The demand distribution is assumed to be Poisson or some modification thereof; the delivery time is assumed to be fixed or Gamma distributed while preserving the sequence of the orders. No demand is lost. It is shown that the demand distribution for the delivery period is negative binomial. Formulae are derived for the re-ordering point and the order quantity and these are solved for the geometric demand distribution. An example is calculated which is based on the stocking of repair parts by a can manufacturer in Chicago.     

变换核估计在未知需求分布函数条件下多阶段库存控制策略中的应用研究

传统的多阶段库存控制主要致力于库存持有以及过多库存的经济性研究.随机库存模型经常假定需求分布已知,这样可以产生容易解决的方案.但随着销售信息的不断更新,需求分布函数的参数常常未知.这样传统的多阶段库存模型很难产生最优的库存控制策略.当前文献对未知需求分布函数条件下的多阶段库存管理问题研究得不多,当需求分布函数随时间变化,是个多阶段随机规划问题,通常情况难以直接进行求解.针对一般非平稳需求,还缺少有效的库存管理方法.本文致力于变换核估计和优化理论相结合的方法研究未知需求分布函数条件下多阶段库存控制策略,提供一条多阶段库存控制的新思路.可以很好地确定各阶段的最优订货点、最高库存、最低库存等来达到整个系统的最优,从而节省更多的成本,达到营运资本的永久性减少、更高的销售量和客户满意度,从而增加企业的竞争力.     

Postponed two-pricing and ordering opportunity for selling a single season inventoried product

Postponement strategies are becoming increasingly important in light of a global trend in which products’ life-cycles are decreasing, such that even products that are not traditionally considered seasonal become “obsolete” within a short period of time (e.g., electronic devices, new cars). Our work addresses postponed-pricing and ordering decisions for a retailer who sells a newsvendor-type inventoried product, in a selling season that is divided into two sub-periods. The division of the selling season enables the retailer to on-line adjust her decisions when faced with a scenario (one that is highly prevalent in reality) in which potential demand changes (increases or decreases) following consumers’ experiences of the product in early stages of the selling season. We assume that the retailer has two opportunities for receiving shipments: prior to the first sub-period and prior to the second one. The retailer determines each order quantity (base-stock level) on the basis of the demand distribution for the corresponding sub-period. In each sub-period, after observing additional market signals, the retailer determines the price of the product for that sub-period. With the aid of a stochastic programming approach, we develop optimization problems and solution methods in order to obtain pricing and ordering decisions that maximize the expected profit of the retailer. We present an extensive numerical example that compares the suggested strategy to three alternative strategies, and conclude that price postponement and responsiveness to demand changes can each reduce leftovers and lost sales as well as substantially increase expected profit.     

Conditions that cause risk pooling to increase inventory

We say product A is a partial substitute for product B if a fraction of the customers who prefer B are willing to accept A when B is out of stock. When demand is uncertain, it is intuitive and true that a larger “willing to substitute” fraction implies larger expected profits. A higher “willing to substitute” fraction allows one to pool the risk of individual products. It may also be intuitive that a larger “willing to substitute” fraction might result in lower optimal total inventory. For the full substitution structure, several researchers have shown that for certain distributions such as the exponential, this latter intuition is not true. We show that this full substitution anomaly can occur with any right skewed demand distribution. We assume i.i.d. demand distributions unless we indicate otherwise. We also show that the anomaly can occur for a number of realistic situations of partial substitution with commonly used demand distributions such as Normal, exponential, Poisson, and uniform. We also demonstrate the anomaly for more than one period, with backlogging, lost sales, more than two products, and with setup costs.