Robust distribution planning for supplier-managed inventory agreements when demand rates and travel times are stationary

Supplier-managed inventory (SMI) is a partnering agreement between a supplier and his customers. Under this SMI agreement, inventory monitoring and ordering responsibilities are entirely transferred to the supplier. Subsequently, the supplier decides both the quantity and timing of his customer deliveries. The inventory routing problem is an underlying optimization model for SMI partnerships to cost-effectively coordinate and manage customer inventories and related replenishments logistics. This paper discusses the case where customer demand rates and travel times are stochastic but stationary, and proposes a version of the inventory routing optimization model that generates optimal robust distribution plans. The approach proposed to obtain and deploy these robust plans combines optimization and Monte Carlo simulation. Optimization is used to determine the robust distribution plan and simulation is used to fine-tune the plan's critical parameters such as replenishment cycle times and safety stock levels. Results of a simplified real-life case implementing the proposed optimization-simulation approach are shown and discussed in detail.     

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.     

The effect of information sharing on supply chain stability and the bullwhip effect

This paper analyzes the bullwhip effect in multi-stage supply chains operated with linear and time-invariant inventory management policies and shared supply chain information. Such information includes past order sequences and inventory records at all supplier stages. The paper characterizes the stream of orders placed at any stage of the chain when the customer demand process is known and ergodic, and gives an exact formula for the variance of the orders placed. The paper also derives robust analytical conditions, based only on inventory management policies, to predict the presence of the bullwhip effect and bound its magnitude. These results hold independently of the customer demand. The general framework proposed in this paper allows for any inventory replenishment policies, any ways of sharing and utilizing information, and any customer demand processes. It is also shown as a special case that sharing customer demand information across the chain significantly reduces, but does not completely eliminate, the bullwhip effect.     

Understanding supply chain dynamics: A chaos perspective

Variability in orders or inventories in supply chain systems is generally thought to be caused by exogenous random factors such as uncertainties in customer demand or lead time. Studies have shown, however, that orders or inventories may exhibit significant variability even if customer demand and lead time are deterministic. In this paper, we investigate how this class of variability, chaos, may occur in a multi-level supply chain and offer insights into how to manage relevant supply chain factors to eliminate or reduce system chaos. The supply chain is characterized by the classical beer distribution model with some modifications. We observe the supply chain dynamics under the influence of various factors: demand pattern, ordering policy, demand-information sharing, and lead time. Through proper decision-region formation, the effect of various factors on system chaos is investigated using a factorial design. The degree of system chaos is quantified using the Lyapunov exponent across all levels of the supply chain. This study shows that, to reduce the degree of chaos in the supply chain system, the adjustment parameters for both inventory and supply line discrepancies should be more comparable in magnitude. Counter-intuitively, in certain decision regions, sharing demand information can do more harm than good. Similar to the bullwhip effect observed previously in demand, we discover the phenomenon of “chaos-amplification” in inventory across supply chain levels.     

Coordinating lead times and safety stocks under autocorrelated demand

We consider a supply chain in which orders and lead times are linked endogenously, as opposed to assuming lead times are exogenous. This assumption is relevant when a retailer’s orders are produced by a supplier with finite capacity and replenished when the order is completed. The retailer faces demands that are correlated over time – either positively or negatively – which may, for example, be induced by a pricing or promotion policy. The auto-correlation in demand affects the order stream placed by the retailer onto the supplier, and this in turn influences the resulting lead times seen by the retailer. Since these lead times also determine the retailer’s orders and its safety stocks (which the retailer must set to cover lead time demand), there is a mutual dependency between orders and lead times. The inclusion of endogenous lead times and autocorrelated demand represents a better fit with real-life situations. However, it poses some additional methodological issues, compared to assuming exogenous lead times or stationary demand processes that are independent over time. By means of a Markov chain analysis and matrix analytic methods, we develop a procedure to determine the distribution of lead times and inventories, that takes into account the correlation between orders and lead times. Our analysis shows that negative autocorrelation in demand, although more erratic, improves both lead time and inventory performance relative to IID demand. Positive correlation makes matters worse than IID demand. Due to the endogeneity of lead times, these effects are much more pronounced and substantial error may be incurred if this endogeneity is ignored.     

Benefits of co-operation in a production distribution environment

We model co-operation in a typical production distribution setting that contains one capacitated supplier producing and distributing a single product to many identical retailers who are facing i.i.d. end-item demands from the consumers. We consider three inventory allocation mechanisms, representing varying degrees of co-operation, at the supplier: (1) the orders from the retailers are filled in a predetermined sequence; (2) the orders from the retailers are filled after taking into account their current inventories; and (3) the orders from the retailers are filled assuming that the product can also be shipped from one retailer to another. We estimate the benefits due to co-operation in this supply chain and study the effect of various system parameters on these benefits. An extensive computational study indicated that the benefits of co-operation in this production distribution environment decrease with increase in the supplier capacity, increase in the number of retailers, decrease in penalty cost, and decrease in consumer demand variance.     

Impact of equity and equality on stability and collusion in a decentralized network

We consider a firm that procures a product from a regular supplier whose production is subject to both supply disruption and random yield risks and a backup supplier whose production capacity requires reservation in advance. Under both deterministic and stochastic demand, we study the impact of the two types of supply risks on the firm’s optimal procurement decisions and the importance of correctly identifying the source of supply risks. We find that if the overall supply risk is unchanged but its main source shifts from random yield to supply disruption, the firm should order more from the regular supplier and reserve less capacity from the backup supplier. Ignoring the existence of supply disruption leads to under-utilization of the regular supplier and over-utilization of the backup supplier. Moreover, we examine the option value of the reserved capacity that is affected by the uncertainty of customer demand. We find that the option value increases/decreases in demand uncertainty if the reservation capacity is exercised after/before demand is realized.     

Mitigating supply risk: an approach with quantity flexibility procurement

This paper considers a buyer that procures from its major supplier whose production is subject to random yield risk. To mitigate supply risk, the buyer can procure from another reliable supplier who provides quantity flexibility (QF) contract. Under both deterministic and stochastic demand, we study the buyer’s optimal procurement decisions. We analyze the structural properties of optimal solutions and identify the conditions under which the quantity flexibility procurement policy should be used. We also examine the effect of supply risk, flexibility, wholesale price and demand risk on the procurement decisions. We find that the higher supply risk and demand risk reduce the buyer’s profit but have different impact on the buyer’s order policy. For the QF supplier, it may not obtain more orders by providing larger flexibility to the buyer, on the contrary, doing this may benefit the risky supplier. For the QF supplier or risky supplier, given its competitor’s wholesale price, it can increase its order share by lower wholesale price.     

Lot-sizing two-echelon assembly systems with random yields and rigid demand

We consider a two-echelon assembly system producing a single final product for which the demand is known. The first echelon consists of several parallel stages, whereas the second echelon consists of a single assembly stage. We assume that the yield at each stage is random and that demand needs to be satisfied in its entirety; thus, several production runs may be required. A production policy should specify, for each possible configuration of intermediate inventories, on which stage to produce next and the lot size to be processed. The objective is to minimize the expected total of setup and variable production costs.We prove that the expected cost of any production policy can be calculated by solving a finite set of linear equations whose solution is unique. The result is general in that it applies to any yield distribution. We also develop efficient algorithms leading to heuristic solutions with high precision and, as an example, provide numerical results for binomial yields.     

Direct selling by suppliers improves system-wide information flow

Consider a bilateral monopoly selling to a market with uncertain demand. The retailer has access to a demand signal. The supplier can add a direct channel, which grants it market access as well. The supplier and the retailer can acquire signals from each other with payments. We show that direct selling by the supplier improves information flow to realize system-wide information transparency, which has mixed effects on the profits for the retailer and the system.     

A supply chain under limited-time promotion: The effect of customer sensitivity

We consider a two-echelon supply chain with a supplier and a retailer facing stochastic customer demands. The supplier is a leader who determines a wholesale price. In response, the retailer orders products and sets a price which affects customer demands. The goal of both players is to maximize their profits. We find the Stackelberg equilibrium and show that it is unique, not only when the supply chain is in a steady-state but also when it is in a transient state induced by a supplier’s promotion. There is a maximum length to the promotion, however, beyond which the equilibrium ceases to exist. Moreover, if customer sensitivity increases, then the wholesale equilibrium price decreases, product orders increase and product prices drop. This effect, well-observed in real life, does not, however, necessarily imply that the promotion is always beneficial. Conditions for the profitability of a limited-time promotion are shown and analyzed numerically. We discuss both open-loop and feedback policies and derive the conditions necessary for them to remain optimal under stochastic demand fluctuations.     

历史订单信息对牛鞭效应的影响分析

客户需求信息的失真是导致牛鞭效应存在的原因,基于零售商的历史订单数据对其需求进行预测可以部分消除牛鞭效应。论文基于零售商-分销商二级供应链视角,分析了在零售商的需求为线性自回归模式的二级供应链中,分销商利用零售商历史订单数据和现有订单数据进行需求预测时自身库存成本的变更以及整个供应链的牛鞭效应的缓解程度。结果表明:分销商利用历史订单数据进行库存的决策可以显著地降低自己的平均库存和需求的波动,这种降低程度在零售商的订货提前期较大的情况下比较明显,但是零售商的需求预测相关系数对它影响不大。     

The bullwhip effect in supply chain networks

This paper analyzes the propagation and amplification of order fluctuations (i.e., the bullwhip effect) in supply chain networks operated with linear and time-invariant inventory management policies. The supply chain network is allowed to include multiple customers (e.g., markets), any network structure, with or without sharing information. The paper characterizes the stream of orders placed by any supplier for any stationary customer demand processes, and gives exact formulas for the variance of the orders placed and the amplification of order fluctuations. The paper also derives robust analytical conditions, based only on inventory management policies, to predict the presence of the bullwhip effect for any network structure, any inventory replenishment policies, and arbitrary customer demand processes. Numerical examples show that the analytical results accurately quantify the bullwhip effect; managerial insights are drawn from the analysis. The methodology presented in this paper generalizes those in previous studies for serial supply chains.     

Semiconductor lot allocation using robust optimization

In this work, the problem of allocating a set of production lots to satisfy customer orders is considered. This research is of relevance to lot-to-order matching problems in semiconductor supply chain settings. We consider that lot-splitting is not allowed during the allocation process due to standard practices. Furthermore, lot-sizes are regarded as uncertain planning data when making the allocation decisions due to potential yield loss. In order to minimize the total penalties of demand un-fulfillment and over-fulfillment, a robust mixed-integer optimization approach is adopted to model is proposed the problem of allocating a set of work-in-process lots to customer orders, where lot-sizes are modeled using ellipsoidal uncertainty sets. To solve the optimization problem efficiently we apply the techniques of branch-and-price and Benders decomposition. The advantages of our model are that it can represent uncertainty in a straightforward manner with little distributional assumptions, and it can produce solutions that effectively hedge against the uncertainty in the lot-sizes using very reasonable amounts of computational effort.     

Optimal policies for inventory usage,production and pricing of fashion goods over a selling season

A short selling season and highly uncertain demands prior to the season characterize production and selling of fashion goods. Once the season starts and demands turn up with a peak interest in the beginning, monopoly becomes under tremendous pressure to produce the required amount so as not to disappoint its customers. It motivates the monopoly to prepare significant inventories by the opening day. Unfortunately, even the most advanced techniques for demand forecasting are likely to induce either an overestimate or underestimate of the initial inventories. Both affect the monopoly's profit. Overestimation results in surplus, which may never be sold, and excessive inventory holding costs. Underestimation implies sales as well as customer loyalty losses. Given inventory level at the beginning of the selling season, we derive policies of handling this inventory, production capacity and product prices in order to maximize the profit and thus diminish the effect of inherent inaccuracy of initial inventory estimation of fashion goods. A case of bookstore management illustrates the effectiveness of the suggested strategies.     

Long-term and penalty contracts in a two-stage supply chain with stochastic demand

Recent applications of game-theoretic analysis to supply chain efficiency have focused on constructs between a buyer (the retailer or manufacturer) and a seller (the supplier) in successive stages of a supply chain. If demand for the final product is stochastic then the supplier has an incentive to keep its capacity relatively low to avoid creating unneeded capacity. The manufacturer, on the other hand, prefers the supplier’s capacity to be high to ensure that the final demand is satisfied. The manufacturer therefore constructs a contract to induce the supplier to increase its production capacity. Most research examines contracting when final demand is realized after the manufacturer places its order to the supplier. However, if final demand is realized before the manufacturer places its order to the supplier, these types of contracts can be ineffective. This paper examines two contracts under the latter timing scenario: long-term contracts in which the business relationship is repeated, and penalty contracts in which the supplier is penalized for too little capacity. Results indicate long-term contracts increase the profit potential of the supply chain. Furthermore, the penalty contracts can ensure that the supplier chooses a capacity level such that the full profit potential is achieved.     

Time-dependent and independent control rules for coordinated production and pricing under demand uncertainty and finite planning horizons

We address the effect of uncertainty on a manufacturer’s dynamic production and pricing decisions over a finite planning horizon. The demand for products, which depends on their price, is characterized by two stochastic processes: potential demand and customer price sensitivity. An optimal policy for coordinating production and pricing is a time-dependent feedback rule with respect to the state of the manufacturer’s inventories. We show that when the volatility of customer sensitivity to the product price is negligible, the optimal policy can be obtained analytically. Moreover, our simulations demonstrate that the volatility of stochastic customer price sensitivity does not have a strong effect on the manufacturer’s expected profit. Therefore, the solution derived for the case of customer price sensitivity with zero volatility can serve as a good approximation heuristic for the optimal policy if the true volatility of customer price sensitivity is within 40 % of its mean and the volatility of potential demand is within 25 % of its mean. Moreover, under these conditions, a simplified, time-independent control rule deteriorates expected profits by only 1.5 %.     

The scheduling of deliveries in a production-distribution system with multiple buyers

In a real production and distribution business environment with one supplier and multiple heterogeneous buyers, the differences in buyers’ ordering cycles have influence on production arrangements. Consequently, the average inventory level (AIL) at the supplier’s end is affected by both the production policy and the ordering policy, typically by the scheduling of deliveries. Consequently, the average inventory holding cost is most deeply affected. In this paper, it is proposed that the scheduling of deliveries be formulated as a decision problem to determine the time point at which deliveries are made to buyers in order to minimize the supplier’s average inventory. A formulation of the average inventory level (AIL) in a production cycle at the supplier’s end using a lot-for-lot policy is developed. Under the lot-for-lot policy, the scheduling of deliveries (SP) is formulated as a nonlinear programming model used to determine the first delivery point for each buyer with an objective to minimize the sum of the product of the individual demand quantity and the first delivery time for each buyer. Thus, the SP model determines not only the sequence of the first deliveries to individual buyers, but also the time when the deliveries are made. An iterative heuristic procedure (IHP) is developed to solve the SP model assuming a given sequence of buyers. Six sequence rules are considered and evaluated via simulation.     

A queueing model of delayed product differentiation

We consider a production system in which a supplier produces semi-finished items on a make-to-stock basis for a manufacturer that will customize the items on a make-to-order basis. The proportion of total processing time undertaken by the supplier determines how suitable the semi-finished items will be to meet customer demand. The manufacturer wishes to determine the optimal point of differentiation (the proportion of processing completed by the supplier) and its optimal semi-finished goods buffer size. We use matrix geometric methods to evaluate various performance measures for this system, and then, with enumeration techniques, obtain optimal solutions. We find that delayed product differentiation is attractive when the manufacturer can balance the costs of customer order fulfillment delay with the costs associated with unsuitable items.     

Component ordering strategies in assembly systems with uncertain capacity and random yield

Sourcing components in a complex global supplier network may lead to a high degree of supply uncertainty. Events, such as unexpected production defects or insufficient supplier capacity, can cause unexpected shortages of required components and halt the assembly of final products. Accordingly, the assembly enterprises must effectively manage various supply uncertainties in their component ordering decisions to avoid such component shortfalls. These issues have guided this research to investigate the optimal ordering strategies of an assembler facing the following two types of supply uncertainty: the uncertain production capacity of a standard component (component 1) and the random production yield of a core component (component 2). The assembler makes the component ordering decisions before these supply uncertainties are realized. We characterize the optimal ordering decision and find that the assembler should order components 1 and 2 according to a fixed ratio, which only depends on the random yield of component 2 and the production cost of component 1, but not on the uncertain capacity of component 1. A case study is presented to further explore the intertwined effects of these two uncertainties in an assembly system. Finally, the model is extended to consider a secondary option of buying additional component 1 s after observing some or all of the supply uncertainties, and this secondary option endows the firm with different capabilities in counteracting the supply uncertainties.