218. The effect of demand sharing behavior on the bullwhip effect in the semiconductor industry
Invited abstract in session TA-8: Technology, Trust, and Tensions: Strategic Behavior in Supply Chains, stream Game Theory and Behavioral Management Science.
Thursday, 8:45-10:15Room: H8
Authors (first author is the speaker)
| 1. | Isabella Lippert
|
| 2. | Hans Ehm
|
| Supply Chain, Infineon | |
| 3. | David Wuttke
|
| TUM Campus Heilbronn, Technical University of Munich, TUM School of Management, TUM Campus Heilbronn |
Abstract
Supply chains continue to suffer from the bullwhip effect, where demand variability is amplified as orders move upstream. Despite decades of research and operational improvements, this phenomenon remains persistent and disruptive. Its impact has been severe in the semiconductor industry, where inherent long production times magnify the consequences. Recent disruptions have revived calls for improved coordination mechanisms, including initiatives to share "true" demand across tiers of the supply chain.
Demand sharing appears to be a promising solution: accurate demand forecasts from downstream firms could help upstream firms plan more effectively, reducing volatility and inefficiencies.
Does such information sharing work in practice? More critically, do downstream firms share truthfully? In industries like semiconductors, characterized by scarce capacity, long lead times, and volatile demand, OEMs may strategically inflate forecasts to secure future capacity or buffer against uncertainty. Can strategies such as introducing an additional anonymous forecast overcome these tactically induced shortfalls?
This paper examines whether demand sharing mitigates the bullwhip effect and/or it becomes a new arena for strategic behavior. We investigate this question through three hypotheses, focusing on how downstream firms report demand and how this affects upstream dynamics. To test these hypotheses, we adapt the MIT Beer Game to capture the unique structure of semiconductor supply chains, modifying the number of tiers, the lead time configuration, and the information environment. Most notably, we enable downstream firms to share demand forecasts with the rest of the supply chain, while allowing them to deviate from the truth if they believe it serves their interests.
Keywords
- Supply Chain Management
- Game Theory
- Forecasting applications
Status: accepted
Back to the list of papers