Guillaume Roels

Regret in the Newsvendor Model with Partial Information

Traditional stochastic inventory models assume full knowledge of the demand probability distribution. However, in practice, it is often difficult to completely characterize the demand distribution, especially in fast-changing markets. In this paper, we study the newsvendor problem with partial information about the demand distribution (e.g., mean, variance, symmetry, unimodality). In particular, we derive the order quantities that minimize the newsvendors maximum regret of not acting optimally. Most of our solutions are tractable, which makes them attractive for practical application. Our analysis also generates insights into the choice of the demand distribution as an input to the newsvendor model. In particular, the distributions that maximize the entropy perform well under the regret criterion. Our approach can be extended to a variety of problems that require a robust but not conservative solution.

The Price of Anarchy in Supply Chains: Quantifying the Efficiency of Price-Only Contracts

In this paper, we quantify the efficiency of decentralized supply chains that use price-only contracts. With a price-only contract, a buyer and a seller agree only on a constant transaction price, without specifying the amount that will be transferred. It is well known that these contracts do not provide incentives to the parties to coordinate their inventory/capacity decisions. We measure efficiency with the price of anarchy (PoA), defined as the largest ratio of profits between the integrated supply chain (that is, fully coordinated) and the decentralized supply chain. We characterize the efficiency of various supply chain configurations: push or pull inventory positioning, two or more stages, serial or assembly systems, single or multiple competing suppliers, and single or multiple competing retailers.

Contracting for Collaborative Services

In this paper, we analyze the contracting issues that arise in collaborative services, such as consulting, financial planning, and information technology outsourcing. In particular, we investigate how the choice of contract type---among fixed-fee, time-and-materials, and performance-based contracts---is driven by the service environment characteristics. We find that fixed-fee contracts contingent on performance are preferred when the service output is more sensitive to the vendors effort, that time-and-materials contracts are optimal when the output is more sensitive to the buyers effort, and that performance-based contracts dominate when the output is equally sensitive to both the buyers and the vendors inputs. We also discuss how the performance of these contracts is affected with output uncertainty, process improvement opportunities, and the involvement of multiple buyers and vendors in the joint-production process. Our model highlights the trade-offs underlying the choice of contracts in a collaborative service environment and identifies service process design changes that improve contract efficiency.

Robust Controls for Network Revenue Management

Revenue management models traditionally assume that future demand is unknown but can be described by a stochastic process or a probability distribution. Demand is, however, often difficult to characterize, especially in new or nonstationary markets. In this paper, we develop robust formulations for the capacity allocation problem in revenue management using the maximin and the minimax regret criteria under general polyhedral uncertainty sets. Our approach encompasses the following open-loop controls: partitioned booking limits, nested booking limits, displacement-adjusted virtual nesting, and fixed bid prices. In specific problem instances, we show that a booking policy of the type of displacement-adjusted virtual nesting is robust, both from maximin and minimax regret perspectives. Our numerical analysis reveals that the minimax regret controls perform very well on average, despite their worst-case focus, and outperform the traditional controls when demand is correlated or censored. In particular, on real large-scale problem sets, the minimax regret approach outperforms by up to 2% the traditional heuristics. The maximin controls are more conservative but have the merit of being associated with a minimum revenue guarantee. Our models are scalable to solve practical problems because they combine efficient (exact or heuristic) solution methods with very modest data requirements.

An Analytical Model for Traffic Delays and the Dynamic User Equilibrium Problem

In urban transportation planning, it has become critical (1) to determine the travel time of a traveler and how it is affected by congestion, and (2) to understand how traffic distributes in a transportation network. In the first part of this paper, we derive an analytical function of travel time, based on the theory of kinematic waves. This travel-time function integrates the traffic dynamics as well as the effects of shocks. Numerical examples demonstrate the quality of the analytical function, in comparison with simulated travel times. In the second part of this paper, we incorporate the travel-time model within a dynamic user equilibrium (DUE) setting. We prove that the travel-time function is continuous and strictly monotone if the flow varies smoothly. We illustrate how the model applies to solve a large network assignment problem through a numerical example.

Optimal Design of Social Comparison Effects: Setting Reference Groups and Reference Points

In this paper, we study how social planners should exploit social comparisons to pursue their objectives. We consider two modes of social comparison, referred to as behind-averse and ahead-seeking behaviors, depending on whether individuals experience a utility loss from underperforming or a utility gain from overperforming relative to their peers. Modeling social comparison as a game between players, we find that ahead-seeking behavior leads to output polarization, whereas behind-averse behavior leads to output clustering. A social planner can mitigate these effects in two ways: i by providing the full reference distribution of outputs instead of an aggregate reference point based on the average output and ii by assigning players into uniform rather than diverse reference groups. Social planners may thus need to tailor the reference structure to the predominant mode of social comparison and their objective. A performance-focused social planner may set the reference structure so as to maximize the output of either the top or the bottom player depending on whether she puts greater marginal weight to larger or smaller outputs. When the social planner also cares about utility, she faces a dilemma because performance optimization may not be aligned with utility maximization. Inevitably, the social planner will have to confront equity issues because better performance may not reflect greater effort or greater ability. This paper was accepted by Serguei Netessine, operations management.

Optimal Design of Coproductive Services: Interaction and Work Allocation

In services, customers provide significant inputs into the production process. In particular, these inputs may be the customers themselves participating in the service delivery. Although many service firms have explored different ways of involving customers in their production process, there is no clear guideline for the design of such coproductive systems. In this paper, we develop an analytical model of joint production between a service provider and a customer and characterize how a service firm should design its coproductive system. We show that, as a task becomes more standard, it is desirable to decrease the degree of interaction between the provider and the customer by making their efforts more substitutable and to allocate most of the work to whoever is the most efficient. Conversely, as a task becomes less standard, it is optimal to increase interaction by making efforts more complementary and to balance the work allocation. Our analysis gives rise to a service-process framework with three archetypes of coproductive services: collaborative services, service factories, and self-services. We discuss the implications of our results for service process reengineering.

Win-Win Capacity Allocation Contracts In Co-Production and Co-Distribution Alliances

In some strategic alliances, a firm shares its manufacturing capacity with another, and the latter shares its distribution capacity with the former. Even though such bidirectional alliances have become common, they remain challenging to manage due to the frequent disputes over capacity allocation especially when demand is uncertain. In this paper, we investigate whether there exists a contractual mechanism that can mitigate the extent of these disputes while improving the profits of all participating firms. We consider two types of bidirectional contracts, namely, the ex-post transfer payment contract and the ex-ante capacity reservation contract. By modeling the capacity allocation and the bidirectional contract design as a multi-stage game between two firms with non-competing product lines, we show that, relative to a situation where capacity allocation is freely negotiated, either contract can improve the alliance’s total profit in equilibrium. In terms of distribution of the total surplus, we find that capacity reservation contracts always make both firms better off, whereas ex-post transfer payment contracts may make one firm worse off. Hence, capacity reservation contracts are more likely to be implemented in practice in such co-production and co-distribution alliances.

Risk Premiums in the Price-Setting Newsvendor Model

Consider a make-to-stock firm that needs to decide its price and order quantity before a selling season without knowing the demand. By charging a different price from the riskless price, the firm can hedge its profits against demand uncertainty either by mitigating its demand risk or by lowering its opportunity cost of stockouts. In this paper, we characterize the behavior of the risk premium under an additive-multiplicative demand model. We show that the sign of the risk premium can be uniquely determined by the elasticities of the mean and of the standard deviation of demand at the riskless price. We also quantify the worst-case magnitude of the risk premium and find that, although the risk premium can be significant, its range of potential values rapidly narrows and becomes relatively insensitive to the specific elasticity values as the mean demand becomes more elastic. Our characterization of the risk premium generates insights into the role of pricing as a hedge against demand uncertainty.

Team Leadership and Performance: Combining the Roles of Direction and Contribution

In knowledge-intensive projects, one of the challenges project team leaders often face is how to combine their roles of direction and contribution. In this paper, we propose a game-theoretic model of team leadership of coproductive projects and study how team leaders should combine their directing and contributing efforts depending on the team and project characteristics. Our analysis reveals that two types of team leadership approaches arise in equilibrium, namely, “participatory” team leadership, under which the team leader gives the team members full discretion on their choice of effort, and “directive” team leadership, under which the team leader demands team members exert higher effort than what they would choose to exert voluntarily. We find that directive team leadership is optimal when the team members have low incentives, that is, when their rewards are low, the size of the team is large, or failure is not too costly (e.g., continuation is possible); otherwise, participatory team leadership is op...