Skip to main content

Enterprise-Driven Decision-Based Design

Under the Decision-based Design (DBD) framework, engineering design is modeled as a decision-making process that seeks to maximize the value of a designed artifact under uncertainty and risk. At the core of this research area is the establishment of rigorous decision-making principles, and the study of the interrelationships between enterprise product planning and engineering product development. Currently, our research focuses on developing analytical approaches that bridge the gap between quantitative engineering design decisions and qualitative customer preferences through demand modeling.

Books and Papers on Decision-based Design

Jump to

Network-based Customer Preference Modeling

Our current work in demand modeling is focused on modeling the heterogeneity of customer preferences and the impact of social networks on customer choices. In the proposed Multidimensional Customer-Product Network (MCPN), customer-product interactions are viewed as a socio-technical system where separate entities of “customers” and “products” are simultaneously modeled, and multiple types of relations, such as consideration and purchase, product associations, and customer social connections are considered. Beyond the traditional descriptive analysis that examines the network structure characteristics, we employ the Exponential Random Graph Model (ERGM) as a unified statistical inference framework to interpret complex preference decisions. Our approach broadens the traditional utility-based logit models by considering the dependency among complex customer-product relations, the “irrationality” of customers induced by social influence, nested multi-choice decisions, and correlated attributes of customers and products. The methods developed will enhance industry’s resilience to changing markets through strategic product design. Our design framework allows designers to focus their engineering efforts on the most critical feature upgrades and quality improvements, considering not only the engineering requirements but also the business interests, customers’ desires, and market behavior.

Representative papers

Multilevel Decision-based Design (DBD)

DBD is a collaborative multidisciplinary design approach that employs a single value function (e.g., profit) from the enterprise’s perspective to maximize the expected utility of a designed artifact while also considering uncertainty and the decision maker’s risk attitude. The core of enterprise-driven design is the use of demand modeling to estimate the effect of design changes on a product’s market share, aggregated from individual customers’ choice probabilities, and consequently on the firm’s revenues. Our research in this area has progressed from the multinomial logit approach for a single product design, to the nested-logit approach for platform-based product family design, to the use of a Bayesian framework for supporting complex hierarchical systems design, and to use of choice modeling in usage and social context-based design. These approaches have been applied in various practical design problems in industry, such as the universal motor design, vehicle engine design, suspension systems design, and vehicle occupant package design. In addition, to manage the complexity of DBD, we developed a multilevel DBD framework where the product planning from the enterprise level identifies the optimal settings of engineering design attributes that serve as the targets to be optimized at the product level. Compared to the all-in-one single-objective DBD approach, the multilevel optimization approach is more representative of the hierarchical decision-making used in a typical industry design process.

Representative papers

List of Publications

Copyright 2024 Integrated DEsign Automation Laboratory (IDEAL) at Northwestern University. All Rights Reserved.