Modular product systems design and planning: Assessing the impact of configuration flexibility

This thesis examines product-line design issues for modular product systems. Modular products consist of subassemblies connected by interfaces that permit product configurations to evolve over time via module substitutions. The configuration flexibility provided by interchangeable modules enables consumers to substitute individual modules. The significance of this mix-and-match flexibility must be addressed by firms when making product-line planning decisions. Existing product-line design models for integrative products don't address the configuration flexibility aspect of modular systems. This thesis provides frameworks for assessing the value of modularity and for planning the optimal set of module varieties to offer in a modular product line.;To assess the value of the asynchronous module interchanges made feasible by modular configurations, we develop a long-term model in which product technologies develop at distinct rates. If the technologies underlying different product subassemblies develop at diverse rates, then producing those subassemblies as distinct modules permits consumers to derive increased surplus over time, and some of that surplus may be extracted by the firm. These benefits must be weighed against increases in manufacturing costs entailed by modular systems, which sometimes result from the need to replicate subsystems and provide robust interfaces. Our long-term model of modular systems contrasts the profitability of modular and integrative designs by analyzing this manufacturing cost differential and the value of enabling consumers to upgrade individual modules over time. An extension to this model considers the ability of modularity to provide a barrier-to-entry.;We also present a modular product-line planning model that considers a single generation of a modular system, for the purpose of determining the optimal set of module varieties to manufacture. Customer switching between modules is analyzed, and consumer uncertainty is explicitly modeled as a stochastic discrete-choice problem. Module varieties may be produced using common components as a means of reducing fixed manufacturing costs, subject to quality constraints governing component substitutions. Expected demands for products and modules are derived by deriving customer purchase probabilities. Minimum-cost manufacturing plans are generated via dynamic programming. The optimal set of module varieties to offer is determined by optimizing over all feasible module sets.