Session: DAC-09-1: Design of Complex Systems

Paper Number: 143633

143633 - Modeling Demand Spikes in Continuously Evolving Products Using Dynamic Customer Preferences 

As the digital marketplace continues to evolve with rapid changes in consumer preferences and the frequent introduction of product updates, traditional models for estimating demand are often unsuitable for use. These models often fail to account for the dynamic nature of product development and the complex decision-making processes of consumers, particularly in the face of product updates. To bridge this gap, our study extends a demand model based on decision field theory (DFT), focusing on a phenomenon largely unexplored in the existing literature: the update engagement spike (UES). Being able to capture this phenomenon, where demand surges following a product update, is crucial for accurately modeling demand around product updates. We extend the DFT-based demand model to capture consumer responses to updates more effectively. Our model integrates the dynamics of product updates into the decision-making process by adjusting the decision threshold, effectively capturing the increased likelihood of consumers trying a product post-update. We then validate this extended model using real-world data from competing software products, demonstrating its practical use. Furthermore, we outline avenues for future research, including methods for distinguishing between customer segments and attributes with greater accuracy. To improve the model's applicability and practical use, future work will focus on differentiating between attributes and customer segments.

Presenting Author: Ian Walter Purdue University

Presenting Author Biography: Ian Walter is a PhD student in the Elmore Family School of Electrical and Computer Engineering at Purdue University.

Authors:

Ian Walter Purdue University
Phil Pare Purdue University
Jitesh Panchal Purdue University