Strategic Design Decisions for Uncertain Market Systems Using an Agent Based Approach

2011 ◽  
Vol 133 (4) ◽  
Author(s):  
Z. Wang ◽  
S. Azarm ◽  
P. K. Kannan

Market players, such as competing manufacturing firms and retail channels, can significantly influence the demand and profit of a new product. Existing methods in design for market systems use game theoretic models that can maximize a firm’s profit with respect to the product design and price variables given the Nash equilibrium of the market system. However, in the design for uncertain market systems, there is seldom equilibrium with players having fixed strategies in a given time period. In this paper, we propose an agent based approach for design for market systems that accounts for learning behaviors of the market players under uncertainty. By learning behaviors we mean that market players gradually, over time, learn to play with better strategies based on action–reaction behaviors of other players. We model a market system with agents representing competing manufacturers and retailers who possess learning capabilities and based on some prespecified rules are able to react and make decisions on the product design and pricing. The proposed agent based approach provides strategic design and pricing decisions for a manufacturing firm in response to possible reactions from market players in the short and long term horizons. Our example results show that the proposed approach can produce competitive strategies for the firm by simulating market players’ learning behaviors when they react only by setting prices, as compared to a game theoretic approach. Furthermore, it can yield profitable product design decisions and competitive strategies when competing firms react by changing design variables in the short term—case for which no previous method in design for market systems has been reported.

Author(s):  
Z. Wang ◽  
S. Azarm ◽  
P. K. Kannan

Market players, such as competing manufacturing firms and retail channels, can significantly influence the demand and profit of a new product. Existing methods in design for market systems use game theoretic models that can maximize a focal manufacturing firm’s profit with respect to product design and price variables given the Nash equilibrium of the market system. However, in the design for uncertain market systems, there is seldom equilibrium with players having fixed strategies in a given time period. In this paper, we propose an agent based approach for design for market systems that accounts for learning behaviors of the market players under uncertainty. By learning behaviors we mean that market players gradually, over a time period, learn to play with better strategies based on action-reaction behaviors of other players. We model a market system with agents representing competing manufacturers and retailers who possess learning capabilities and are able to automatically react and make decisions on the product design and pricing. The proposed approach provides strategic design and pricing decisions for a focal manufacturer in response to anticipated reactions from market players in the short and long term horizons. Our example results show that the proposed agent based approach can produce competitive strategies for a focal firm over a time period when market players react only by setting prices compared to a game theoretic approach. Furthermore, it can yield profitable product design decisions and competitive strategies when competing firms react by changing design attributes in the short term — a case for which no previous method in design for market systems has been reported.


Author(s):  
Brendan E. Fay ◽  
Steven Hoffenson

In product design, there is often a disconnect between the engineers creating the product and the marketing team determining the best characteristics for the product. The decisions of engineers in designing a product, such as dimensions, materials, and tolerances, affect the look and feel of the product as seen by the consumer. This paper presents a methodology for creating a simulation to demonstrate basic relationships between product design decisions and the predicted success of the product in the market. An agent-based model is used to simulate the interactions among producers and consumers in a market system, using analysis techniques from engineering, economics, and psychology. The effects of design variables on product attributes influence the cost and consumer utility of the product, which, along with price, affect market dynamics. The goal of creating this model is for use in design curricula to introduce the concept of market-driven product design and allow student design teams to assess the impacts of different design variables on the market success of a product. The model has been introduced in an undergraduate engineering design class to explore its effectiveness in teaching market-driven design.


Author(s):  
Ekaterina Sinitskaya ◽  
Kelley J. Gomez ◽  
Qifang Bao ◽  
Maria C. Yang ◽  
Erin F. MacDonald

This work uses an agent-based model to examine how installers of photovoltaic (PV) panels influence panel design and the success of residential solar energy. It provides a novel approach to modelling intermediary stakeholder influence on product design, focusing installer decisions instead of the typical solar stakeholder foci of the final customer (homeowners) and the designer/manufacturer. Installers restrict homeowner choice to a subset of all panel options available, and, consequentially, determine medium-term market dynamics in terms of quantity and design specifications of panel installations. This model investigates installer profit-maximization strategies of exploring new panel designs offered by manufacturers vs. exploiting market-tested technology. Manufacturer design decisions and homeowner purchase decisions are modeled. Realistic details provided from installer and homeowner interviews are included. For example, installers must estimate panel reliability instead of trusting manufacturer statistics, and homeowners make purchase decisions based in part on installer reputation. We find that installers pursue new and more-efficient panels over sticking-with market-tested technology under a variety of panel-reliability scenarios and two different state scenarios (California and Massachusetts). Results indicate that it does not matter if installers are predisposed to an exploration or exploitation strategy — both types choose to explore new panels with higher efficiency.


Author(s):  
Ching-Shin Shiau ◽  
Jeremy Michalek

Recent research has extended prior efforts to integrate firm-level objectives into engineering design optimization models by further enlarging the scope to investigate the effects of regulation on the design decisions of profit-seeking firms in competition. In particular, one study examined the effects of environmental policy on vehicle design decisions by integrating quantitative models of engineering performance, market demand, production cost and regulatory penalties in a joint optimization framework using game theory to model the effects of competition on design and pricing. Model complexity and the solution methods used to solve for market equilibria in prior research have led to a limitation where the prior approach is too computationally intensive to allow extensive parametric studies on the effects of policy changes on design. To address this issue, we present an alternative game-theoretic approach utilizing necessary and sufficient conditions with Nash conditions to find market equilibria in an oligopoly of automakers, and we use this approach to examine the resulting optimal design responses under various regulation scenarios.


Author(s):  
William Steinhour ◽  
K. Krishnamurthy

Abstract To develop a successful product in modern industry, it is necessary to incorporate design performance requirements, manufacturing processes requirements and cost considerations in the early stages of the product design cycle. In this study, a game-theoretic framework for integrated product design is discussed. A minimax game formulation is developed and applied to the design of a simplified automobile door. Results are presented with the design engineer, manufacturing engineer and cost accountant each represented by a player. The minimax game yields a conservative solution and represents the worst-case scenario. This study explores how the game-theoretic approach can be effectively implemented to design a product while satisfying conflicting requirements.


Sign in / Sign up

Export Citation Format

Share Document