Faculty and Research

"Revenue Management in Airline Alliances"

Abstract: Major airlines are selling increasing numbers of interline itineraries, in which flights operated by two or more airlines are combined and sold together. This creates a difficult coordination problem: each member of the alliance makes revenue management decisions to maximize its own revenue, and the resulting behavior may produce sub-optimal revenue for the alliance as a whole. Airline industry researchers and consultants have proposed a variety of static and dynamic mechanisms to control revenue management decisions across alliances (a dynamic mechanism adjusts its parameters as the number of available seats in the network changes). In this talk, we examine Markov-game models of two-partner alliances that can be used to analyze the effects of these mechanisms on each partner's behavior. We begin with models that assume complete information: within each time period, each airline knows its partner's inventory level, and both have identical forecasts of future arrival probabilities and revenue distributions over the entire alliance network. For these games, we show that no Markovian transfer pricing mechanism can coordinate an arbitrary alliance. We then derive equilibrium acceptance policies and generate insights about partner behavior under various coordination schemes. We then consider alliances that operate with incomplete information. We consider an extreme case: that transfer prices are the only information that each airline has about the state of its partner's network. Specifically, its partner's forecasts (revenue and arrival processes) and inventory levels are not shared. Therefore, each partner's decisions are based solely on the transfer prices, their own forecasts for their own networks, and their own inventory levels. We explore the airlines' equilibrium policies under this game and compare the overall performance of the alliance to both the centralized control (first-best) alliance and an alliance that shares complete information.

"Global Dual Sourcing: Mexico or China?" Paper > papers under review> #4

Abstract: When designing a sourcing strategy in practice, a key task is to determine the average order rates placed to each source because that affects cost and supplier management. We consider a firm that has access to a responsive near-shore source (e.g., Mexico) and a low-cost offshore source (e.g., China). The firm must determine an inventory sourcing policy to satisfy random demand over time. Unfortunately, the optimal policy is too complex to allow a direct answer to our key question. Therefore, we analyze a tailored base-surge (TBS) sourcing policy that is simple, used in practice, and captures the classic tradeoff between cost and responsiveness. The TBS policy combines push and pull control by replenishing at a constant rate from the offshore source and producing at the near shore plant only when inventory is below a target. The constant base allocation allows the offshore facility to focus on cost efficiency while the nearshore's quick response capability is utilized only dynamically to guarantee high service. The research goals are to i) determine the allocation of random demand into base and surge capacity, ii) estimate corresponding working capital requirements, and iii) identify and value the key drivers of dual sourcing.

"Implementing and Evaluating Distribution Systems: Experience with Zara and Beyond"

Abstract: Working in collaboration with Spain-based retailer Zara, we address the problem of distributing, over time, a limited amount of inventory across all the stores in a fast-fashion retail network. Challenges specific to that environment include very short product life-cycles, and store policies removing articles from display whenever one of their key sizes stocks out. We first formulate and analyze a stochastic model predicting the sales of an article in a single store during a replenishment period as a function of demand forecasts, the inventory of each size initially available and the store inventory management policy just stated. We then formulate a mixed-integer program embedding a piece-wise linear approximation of the first model applied to every store in the network, allowing us to compute store shipment quantities maximizing overall predicted sales. We report the implementation of this model by Zara, and the ensuing controlled pilot experiment performed to assess its impact relative to the prior procedure used to determine store shipment quantities. This experiment suggests that the new allocation process increases sales by 3 to 4%, which is equivalent to $275M in additional revenues for 2007, reduces transshipments, and increases the proportion of time that Zara's products spend on display within their life-cycle. Zara is currently using this process for all of its products worldwide. If time permits I will also discuss a related ongoing collaboration with the World Bank and Zambia's Ministry of Health to improve the distribution of essential medical drugs in Zambia. (joint work with F. Caro, UCLA; P. Yadav, Zaragoza Logistics Center; Z. Leung, MIT) Paper (attached): "Inventory Management of a Fast-Fashion Retail Network" (with F. Caro), to appear in Operations Research, 2009.

"The Co-Evolution of Organizational Structures and Software Structures in Open Source and Closed Source Projects"

Abstract:This study draws on the product development literature to hypothesize how organizational structures and product design structures co-evolve in open and closed source software projects. We further hypothesize how these structures affect project performance. Findings suggest that organizational structures and software structures become mutually interdependent but also reveal significant differences in the patterns of relationships in Open Source and Closed Source projects. In Open Source, the software structure reflects the organizational structure of its development team whereas in Closed Source the software structure also impacts the organizational structure of the team. Further, the organizational structure impacts productivity in Closed Source projects, while the software structure impacts productivity in Open Source projects. Intriguingly, over time the organizational structure in Closed Source projects evolves to resemble that of Open Source. Our results have important implications for the management and coordination of product development projects.

Keywords: Software Design Structure, Organizational Structure, Social Network, Open Source, Closed Source, Software Evolution; Product Development; Product Design; Organizational Design.

"Are Reservations Recommended?"

Abstract: We examine the role of reservations in capacity-constrained services with a focus on restaurants. Although customers value reservations, restaurants typically neither charge for them nor impose penalties for failing to honor them. However, reservations impose costs on firms offering them. We highlight ways in which reservations can increase a firm's sales by altering customer behavior. First, when demand is uncertain, reservations induce more customers to patronize the restaurant on slow nights. The firm must then trade off higher sales in a soft market with sales lost to no shows on busy nights. Competition makes reservations more attractive as long as enough customers will consider dining at either restaurant. When there are many firms in the market, it is rarely an equilibrium for none to offer reservations. Second, we show that reservations can increase sales by shifting demand from a popular peak period to a less desirable off-peak time. This is accomplished by informing diners when all peak reservations have been given out. In this setting, competition may make offering reservations less attractive, and a market with many firms may have no one offering reservations.

"Managing Customer Goodwill Using Approximate Dynamic Programming"

Abstract: We study the problem faced by a supplier deciding how to dynamically allocate limited capacity among a portfolio of customers who remember the fill rates provided to them in the past when making ordering decisions. Customers differ from one another in their contribution margins, in their sensitivity to the past, and in their demand volatility. We seek to understand the impacts of these factors on the supplier's allocation policy. We approach the problem using approximate dynamic programming, by incorporating a dynamic model of customers' behavioral responses. The model trades-off customer characteristics to rationalize how much goodwill the firm should target for each customer.

"Making Globalization Work for All?: Reflections on Improving Labor Standards in Global Supply Chains"

This talk will examine the conditions under which working conditions and labor standards improve in global supply chains. The research focused on the athletic footwear, apparel, electronics and commodity agriculture (sugar) supply chains of four global brands and consisted of analysis of the audit and sourcing data bases of these companies as well as field research in Brazil, China, Bangladesh, India, Dominican Republic, Honduras, Mexico, Turkey, Vietnam and the United States. Two background papers: one & two

"The Influence of Prior Industry Affiliation on Framing in Nascent Industries: The Evolution of Digital Cameras" (with Mary Benner)

"Trust, Feature Explosion and Simplicity-- Some Cross-Cutting Issues in Computational Market Design"

This talk will examine some of the cross-cutting issues that arise when implementing new forms of electronic markets. A first issue is that of trust and privacy, and is illustrated in the context of clearing of large block portfolio trades in ``upstairs" securities markets. A second issue is that of feature explosion in markets with highly configurable goods, and is illustrated in the context of automated channel abstraction for winner determination in expressive display ad auctions for Internet advertising. A third issue is that of enabling stochastic optimization while retaining straightforward bidding, and is illustrated in the context of dynamic auctions for time-sensitive goods.

http://www.eecs.harvard.edu/~cat/papers/ccse.pdf

http://www.eecs.harvard.edu/econcs/pubs/walsh09.pdf

http://www.eecs.harvard.edu/econcs/pubs/constantinec09.pdf

Clay Christensen HBS

"There aren't as many unique ideas as we think"

Abstract: I'd like to lead a discussion on how particular constructs from our research manifest themselves at different levels of analysis. I'm hoping this will give the faculty a new way of thinking about where to find research opportunities. And if there is time, I'd like to present a couple of new models that have emerged through BSSE (Building and Sustaining a Successful Enterprise) that you may find interesting.

"Does Management Matter? Evidence from India"

A long standing question in social science is whether management matters. Certainly management differs across firms, as does performance. However, perhaps every firm chooses its management practices optimally, so that differences across firms simply reflect differences in their environments. To investigate this we run a field experiment on large Indian textile firms to evaluate the causal impact of modernizing their management practices. We do this by providing free management consulting to a set of randomly chosen treatment plants, and compare their performance to a set of control plants. We find that improved management practices led to significantly higher efficiency and quality, and lower inventory levels. These changes increased the average plant's productivity by about 15% and profitability by about $0.5m (24%) per year. Firms also transferred these improved management practices from their treated plants to other plants within their group. Since firms adopted these profitable management practices this raises the question of why they had not done so before. Our results suggest that informational barriers were initially important in explaining this lack of adoption. Modern management practices are a type of technology that diffuses slowly between firms. In the longer run, constraints around CEO ability and behavior are also important.

Paper: http://mfi.uchicago.edu/programs/fy10_events/papers/feb_26/DMM_Feb2010.pdf

Karim Lakhani HBS

"Open Innovation Institutions as Governance or Selection Mechanisms? Evidence from a Field Experiment on a NASA Space Life Science Problem"

Abstract: In this paper we present novel field experimental evidence to show that individuals have different preferences for cooperative and competitive institutions for innovation---and these preferences have large, economically significant, performance implications. In the experiment, individuals could choose between a competitive and a cooperative regime to solve a computational-engineering problem faced by NASA's Space Life Sciences Directorate. The experiment mimicked "open innovation" institutional regimes where individuals can participate in either competitive or cooperative problem solving with both pecuniary and non-pecuniary reward structures. Here we compare the performance of individuals who chose to be in the competitive regime versus those that were assigned to the regime, controlling for their skill level. First we find that those who were given the choice of institutional regime performed significantly better and exerted more effort than those that were assigned to compete. Second we find that the presence of pecuniary rewards resulted in better performance and more effort. Third we find that the impact of choice on performance and effort is equivalent to- and sometimes greater than the impact of pecuniary rewards, showing the importance of selection and sorting for innovation.

"The cost of carbon and supply chain design"

Reducing greenhouse gas emissions is a significant challenge for our society. The transportation sector is a large contributor to greenhouse gas emissions. Consequently, the design of a supply chain influences its carbon footprint. I will focus on the structure of the retail supply chain and in particular, how will (and should) the design of a retail supply chain change as carbon becomes more expensive.

Deishin Lee HBS

"Waste-to-Energy: Operating Under Regulation"

Abstract: Two compelling environmental challenges of the industrial world today are pollution from waste disposal and emissions from energy consumption. An elegant, viable business solution to these environmental concerns is the conversion of waste into energy. Given the process characteristics of the waste-to-energy operation, the market characteristics for waste disposal and energy, and the mechanisms regulators use to encourage production of renewable energy, we determine the profit-maximizing operating strategy for the waste-to-energy _rm. In one operation, the waste-to-energy firm serves two types of customers: waste generators who pay for waste disposal service, and electricity consumers who buy energy. Our results show that the firm should leverage the service side of the business and accept the entire, potentially volatile amount of waste from the waste generator (operating in "service mode") instead of contracting for a steady baseload fraction of the waste (operating in "procurement mode") even though it increases the cost of energy production by incurring landfill costs. We also show how regulatory subsidies for renewable energy affect the operating decisions of the firm. We find that for the same dollar amount, distributing the subsidy as a lump sum rather than as a price premium per kilowatt-hour of electricity induces the waste-to-energy firm to make operating decisions that are better for the environment.

Melissa Schilling NYU

"When Hubs Forget, Lie, and Play Favorites: Interpersonal Network Structure, Information Distortion, and Organizational Learning"

The interpersonal network structure of an organization directly influences the diffusion and recombination of ideas, and can thus facilitate or impede organizational learning. We examine the influence of a ubiquitous feature of such networks: Hubs. In most interpersonal networks, there are individuals who have significantly more connections than does the average member. A "hubby" network typically enables much faster information diffusion because hubs provide shortcuts between many others in the network. The influence of hubs on organizational learning, however, is more nuanced than this fact might suggest. First, too rapid diffusion of information can thwart long-run learning by enabling seemingly superior ideas to extinguish information variety in the firm. Second, because hubs have many connections, they might be prone to overload causing them to fail to transmit all of their information ("forgetting"). Third, because hubs can exert great influence over whether, and to whom information is transmitted, they might deliberately misrepresent information ("lying") or withhold information from non-preferred contacts ("playing favorites"). We find that moderately hubby networks outperform both very hubby (scale-free) networks and democratic networks. We also find that moderate amounts of information omission or distortion can be beneficial to long-run organizational performance by preserving information diversity.