Özalp Özer

Strategic Commitments for an Optimal Capacity Decision Under Asymmetric Forecast Information

We study the important problem of how to assure credible forecast information sharing between a supplier and a manufacturer. The supplier is responsible for acquiring the necessary capacity before receiving an order from the manufacturer who possesses private forecast information for her end product. We address how different contracts affect the suppliers capacity decision and, hence, the profitability of the supplier and the manufacturer. We fully develop two contracts (and provide explicit formulae) to enable credible forecast information sharing. The first is a nonlinear capacity reservation contract under which the manufacturer agrees to pay a fee to reserve capacity. The second is an advance purchase contract under which the manufacturer is induced to place a firm order before the supplier secures the component capacity used to build the end product. These contracts serve a strategic role in information sharing. The capacity reservation contract enables the supplier to detect the manufacturers private forecast information, while the advance purchase contract enables the manufacturer to signal her forecast information. We show that channel coordination is possible even under asymmetric forecast information by combining the advance purchase contract with an appropriate payback agreement. Through our structural and numerical results we also show that the degree of forecast information asymmetry and the risk-adjusted profit margin are two important drivers that determine supply chain efficiency and which contract to adopt.

Integrating Replenishment Decisions with Advance Demand Information

There is a growing consensus that a portfolio of customers with different demand lead times can lead to higher, more regular revenues and better capacity utilization. Customers with positive demand lead times place orders in advance of their needs, resulting inadvance demand information. This gives rise to the problem of finding effective inventory control policies under advance demand information. We show that state-dependent ( s, S) and base-stock policies are optimal for stochastic inventory systems with and without fixed costs. The state of the system reflects our knowledge of advance demand information. We also determine conditions under which advance demand information has no operational value. A numerical study allows us to obtain additional insights and to evaluate strategies to induce advance demand information.

Dual Sales Channel Management with Service Competition

We study a manufacturer’s problem of managing his direct online sales channel together with an independently owned bricks-and-mortar retail channel, when the channels compete in service. We incorporate a detailed consumer channel choice model in which the demand faced in each channel depends on the service levels of both channels as well as the consumers’ valuation of the product and shopping experience. The direct channel’s service is measured by the delivery lead time for the product; the retail channel’s service is measured by product availability. We identify optimal dual channel strategies that depend on the channel environment described by factors such as the cost of managing a direct channel, retailer inconvenience, and some product characteristics. We also determine when the manufacturer should establish a direct channel or a retail channel if he is already selling through one of these channels. Finally, we conduct a sequence of controlled experiments with human subjects to investigate whether our model makes reasonable predictions of human behavior. We determine that the model accurately predicts the direction of changes in the subjects’ decisions, as well as their channel strategies in response to the changes in the channel environment. These observations suggest that the model can be used in designing channel strategies for an actual dual channel environment. 1

Trust in Forecast Information Sharing

This paper investigates the capacity investment decision of a supplier who solicits private forecast information from a manufacturer. To ensure abundant supply, the manufacturer has an incentive to inflate her forecast in a costless, nonbinding, and nonverifiable type of communication known as “cheap talk.” According to standard game theory, parties do not cooperate and the only equilibrium is uninformative---the manufacturers report is independent of her forecast and the supplier does not use the report to determine capacity. However, we observe in controlled laboratory experiments that parties cooperate even in the absence of reputation-building mechanisms and complex contracts. We argue that the underlying reason for cooperation is trust and trustworthiness. The extant literature on forecast sharing and supply chain coordination implicitly assumes that supply chain members either absolutely trust each other and cooperate when sharing forecast information, or do not trust each other at all. Contrary to this all-or-nothing view, we determine that a continuum exists between these two extremes. In addition, we determine (i) when trust is important in forecast information sharing, (ii) how trust is affected by changes in the supply chain environment, and (iii) how trust affects related operational decisions. To explain and better understand the observed behavioral regularities, we also develop an analytical model of trust to incorporate both pecuniary and nonpecuniary incentives in the game-theoretic analysis of cheap-talk forecast communication. The model identifies and quantifies how trust and trustworthiness induce effective cheap-talk forecast sharing under the wholesale price contract. We also determine the impact of repeated interactions and information feedback on trust and cooperation in forecast sharing. We conclude with a discussion on the implications of our results for developing effective forecast management policies. This paper was accepted by Ananth Iyer, operations and supply chain management.

Information Acquisition for Capacity Planning via Pricing and Advance Selling: When to Stop and Act?

This paper investigates a capacity planning strategy that collects commitments to purchase before the capacity decision and uses the acquired advance sales information to decide on the capacity. In particular, we study a profit-maximization model in which a manufacturer collects advance sales information periodically prior to the regular sales season for a capacity decision. Customer demand is stochastic and price sensitive. Once the capacity is set, the manufacturer produces and satisfies customer demand (to the extent possible) from the installed capacity during the regular sales period. We study scenarios in which the advance sales and regular sales season prices are set exogenously and optimally. For both scenarios, we establish the optimality of a control band or a threshold policy that determines when to stop acquiring advance sales information and how much capacity to build. We show that advance selling can improve the manufacturers profit significantly. We generate insights into how operating conditions (such as the capacity building cost) and market characteristics (such as demand variability) affect the value of information acquired through advance selling. From this analysis, we identify the conditions under which advance selling for capacity planning is most valuable. Finally, we study the joint benefits of acquiring information for capacity planning through advance selling and revenue management of installed capacity through dynamic pricing.

If the Inventory Manager Knew: Value of Visibility and RFID under Imperfect Inventory Information

Today many companies rely on computerized tracking of inventory. Although computerized inventory tracking is generally assumed to be accurate, in reality, actual on-hand inventory deviates from the inventory record, which distorts the replenishment process and compromises good inventory management. Discrepancy between the inventory record and physical stock has mainly three sources: shrinkage, misplacement and transaction errors. Due to these factors, the actual stock levels are not known with certainty. Therefore, the system fails to order when it should or it carries more inventory than necessary. Periodic inventory audits are the most common approach to maintaining inventory record accuracy. In this paper, we consider a finite horizon, single-item periodic review inventory control problem in which inventory records are inaccurate. Inventory errors accumulate until an inventory count is conducted. After the inventory count, the inventory record is reconciled: error is corrected and all misplaced items are returned to inventory. We explicitly model how different error sources lead to inventory discrepancies. First we determine effective replenishment policies that use information about actual inventory movement and error sources and minimize total expected inventory cost. Next we also determine a replenishment policy that accounts for inventory inaccuracy without having complete visibility to inventory and error sources. By comparing the total cost of using these replenishment policies, we quantify the true value of inventory visibility, as well as the value of elimination or reduction of the sources of inventory discrepancies offered by the latest tracking technologies such as RFID.

Promised Lead-Time Contracts Under Asymmetric Information

We study the important problem of how a supplier should optimally share the consequences of demand uncertainty (i.e., the cost of inventory excesses and shortages) with a retailer in a two-level supply chain facing a finite planning horizon. In particular, we characterize a multiperiod contract form, the promised lead-time contract, that reduces the suppliers risk from demand uncertainty and the retailers risk from uncertain inventory availability. Under the contract terms, the supplier guarantees on-time delivery of complete orders of any size after the promised lead time. We characterize the optimal promised lead time and the corresponding payments that the supplier should offer to minimize her expected inventory cost, while ensuring the retailers participation. In such a supply chain, the retailer often holds private information about his shortage cost (or his service level to end customers). Hence, to understand the impact of the promised lead-time contract on the suppliers and the retailers performance, we study the system under local control with full information and local control with asymmetric information. By comparing the results under these information scenarios to those under a centrally controlled system, we provide insights into stock positioning and inventory risk sharing. We quantify, for example, how much and when the supplier and the retailer overinvest in inventory as compared to the centrally controlled supply chain. We show that the supplier faces more inventory risk when the retailer has private service-level information. We also show that a supplier located closer to the retailer is affected less by information asymmetry. Next, we characterize when the supplier should optimally choose not to sign a promised lead-time contract and consider doing business under other settings. In particular, we establish the optimality of a cutoff level policy. Finally, under both full and asymmetric service-level information, we characterize conditions when optimal promised lead times take extreme values of the feasible set, yielding the supplier to assume all or none of the inventory risk---hence the name all-or-nothing solution. We conclude with numerical examples demonstrating our results.

Selling to the “Newsvendor” with a forecast update: Analysis of a dual purchase contract

We consider a supply chain in which a manufacturer sells to a procure-to-stock retailer facing a newsvendor problem with a forecast update. Under a wholesale price contract, the retailer waits as long as she can and optimally places her order after observing the forecast update. We show that the retailer’s wait-and-decide strategy, induced by the wholesale price contract, hinders the manufacturer’s ability to (1) set the wholesale price and maximize his profit, (2) hedge against excess inventory risk, and (3) reduce his profit uncertainty. To mitigate the adverse effect of wholesale price contract, we propose the dual purchase contract, through which the manufacturer provides a discount for orders placed before the forecast update. We characterize how and when a dual purchase contract creates strict Pareto improvement over a wholesale price contract. To do so, we establish the retailer’s optimal ordering policy and the manufacturer’s optimal pricing and production policies. We show how the dual purchase contract reduces profit variability and how it can be used as a risk hedging tool for a risk averse manufacturer. Through a numerical study, we provide additional managerial insights and show, for example, that market uncertainty is a key factor that defines when the dual purchase contract provides strict Pareto improvement over the wholesale price contract.

Mechanism Design for Capacity Planning Under Dynamic Evolutions of Asymmetric Demand Forecasts

This paper investigates the role of time in forecast information sharing and decision making under uncertainty. To do so, we provide a general framework to model the evolutions of forecasts generated by multiple decision makers who forecast demand for the same product. We also model the evolutions of forecasts when decision makers have asymmetric demand information and refer to it as the Martingale Model of Asymmetric Forecast Evolutions. This model helps us study mechanism design problems in a dynamic environment. In particular, we consider a suppliers principals problem of eliciting credible forecast information from a manufacturer agent when both firms obtain asymmetric demand information for the end product over multiple periods. The supplier uses demand information to better plan for a capacity investment decision. When the supplier postpones building capacity and screening the manufacturers private information, the supplier and the manufacturer can obtain more information and update their forecasts. This delay, however, may increase respectively, decrease the degree of information asymmetry between the two firms, resulting in a higher respectively, lower cost of screening. The capacity building cost may also increase because of a tighter deadline for building capacity. Considering all such trade-offs, the supplier has to determine i when to stop obtaining new demand information and build capacity, ii whether to offer a screening contract to credibly elicit private forecast information or to determine the capacity level without information sharing, iii how much capacity to build, and iv how to design the overall mechanism so that both firms benefit from this mechanism. This paper provides an answer to these questions. In doing so, we develop a new solution approach for a class of dynamic mechanism design problems. In addition, this paper provides a framework to quantify the option value of time for a strategic investment decision under the dynamic evolutions of asymmetric forecasts. This paper was accepted by Yossi Aviv, operations management.

Bounds, Heuristics, and Approximations for Distribution Systems

This paper develops simple approximate methods to analyze a two-stage distribution system consisting of one warehouse and multiple retailers with stochastic demand. We consider local and central control schemes. The main ideas are based on relaxing and or decomposing the system into more manageable newsvendor-type subsystems. We also provide bounds on the optimal policy and the optimal expected cost. We show that one of the heuristics is asymptotically optimal in the number of retailers. These results provide practically useful techniques as well as insights into stock-positioning issues and the drivers of system performance.