Demand Management in E-fulfillment


On April 2, <link people niels-agatz _blank>Niels Agatz has defended his PhD thesis entitled “Demand Management in E-fulfillment”. Internet retailers are in a unique position to adjust, in real-time, the product and service offering to the customer and to change the corresponding prices. While this flexibility provides a vast potential for demand management to enhance profitability, exploiting them requires sophisticated decision support. The various interrelated trade-offs between customer preferences, incentives, and operational considerations are much too complex for simple intuition to suffice. The number of quantitative models currently available in the literature to support demand management in e-fulfillment is rather limited.

Agatz’ thesis intends to study in a systematic way the different methods for demand management in e-fulfillment. Using a case study at a large Dutch e-grocer, he identifies relevant planning issues in this context. He mainly focuses on deliveries where the customer needs to be at home at the time of delivery. The web shop can then offer a menu of time slots to coordinate the delivery with the customer. Demand can be steered by offering different time slots at corresponding higher or lower prices, with the objective to enhance revenues and save costs. Both options (time slot and cost) can be deployed statically, on the basis of demand forecasts, as well as dynamically, on the basis of realised orders. This thesis presents different decision support models for time slot management, both forecast-based and in real time. The usability and the advantages of the different methods are evaluated through computer simulations using real-life order data. The results of this study show that a more dynamic and differentiated way of demand management may lead to considerable cost savings and revenue gains for the internet retailers.

Internet retailers are in a unique position to adjust, in real-time, the product and service offering to the customer and to change the corresponding prices. While this flexibility provides a vast potential for demand management to enhance profitability, exploiting them requires sophisticated decision support. The various interrelated trade-offs between customer preferences, incentives, and operational considerations are much too complex for simple intuition to suffice. The number of quantitative models currently available in the literature to support demand management in e-fulfillment is rather limited. In addition, the specific issues related to demand management in this setting are also not completely clear yet. This dissertation focuses on identifying the real-life issue and specific characteristics of demand management in e-fulfillment. In addition, it aims at developing quantitative models to support decision-making in this setting.

About Niels Agatz:
<link people niels-agatz _blank>Niels Agatz was born in 's-Hertogenbosch, the Netherlands, on December 26, 1978. From 1998, Niels studied Industrial Engineering and Management Science at the Eindhoven University of Technology, the Netherlands. In 2001, he went on a five month exchange to the Universitat Politecnica de Catalunya, in Barcelona, Spain. The research for his master thesis was carried out at the `Dienst Vervoer en Ondersteuning', the Dutch Ministry of Justice, where he studied the cost-efficiency of prisoner transport in the Netherlands.

In October 2004, Niels became a PhD candidate at the Department of Decision and Information Sciences of the Rotterdam School of Management Erasmus University. His PhD research was conducted in close collaboration with the home shopping channel of grocery retailer Albert Heijn. The project was part of the Transumo program, a platform of companies, governmental organisations and universities aimed at collectively developing knowledge on sustainable mobility in the Netherlands.

His work has been published in the European Journal of Operational Research, Sloan Management Review and Wall Street Journal. Furthermore, a book chapter has appeared in The Vehicle Routing Problem: Latest Advances and New Challenges. He has presented his research at major international conferences, such as the annual conference of the Production and Operations Management Society (POMS), the European Conference on Operational Research (EURO), the INFORMS Revenue Management & Pricing Conference and the International Workshop of Distribution Logistics (IWDL).

In December 2008, Niels was rewarded a Rubicon fellowship from the Dutch Organisation for Scientific Research (NWO) which enables him to spend 12 months as a postdoctoral researcher at the School of Industrial Engineering of the Georgia Institute of Technology in Atlanta, USA. His research interests are in demand management, vehicle routing and the interface between marketing and operations.

Abstract:
Internet retailers are in a unique position to adjust, in real-time, the product and service offering to the customer and to change the corresponding prices. Although this flexibility provides a vast potential for demand management to enhance profitability, standard practices and models to support the decision makers are lacking as of to date. This thesis aims at closing this gap by systematically investigating demand management approaches in e-fulfillment. We identify relevant novel planning issues through an in-depth case study at a Dutch e-grocer. We focus particularly on attended home delivery, where the Internet retailer applies delivery time slots to coordinate the reception of the purchased goods with the customer. The main levers to manage customer demand in such an environment are the offered time slots and the corresponding delivery fees. The Internet retailer may apply both of these options, slotting and pricing, at different moments in the sales process, either off-line prior to the actual order in-take or real-time as demand unfolds. The thesis presents several decision-support models for time slot management, both forecast-based and in real-time. The computational studies on real-life data demonstrate the viability and the merits of these methods. The results show that a more dynamic and differentiated demand management approach can lead to considerable cost savings and revenue gains.

More Information
Pictures of the Defense
Full Text of the Dissertation