Product and Process Modularity's Effects on Manufacturing Agility and Firm Growth Performance

Abstract

Modularity in product design has been hailed as a way to speed new product development (NPD), to reduce NPD cost, and to enhance customization possibilities for consumers. Modularity in process design may speed new product manufacturing setup times, reduce costs, and enhance the profitability of the lower volumes that customization often entails. However, empirical evidence is scarce that either product or process modularity—individually, jointly, or sequentially—actually produce these or other proposed benefits (e.g., performance growth). This study builds on general modular systems theory (GMST) by examining the theoretical relationship between product and process modularity and the effects of each on firm growth performance. Using structural equation modeling, partial versus complete mediation by manufacturing agility is also scrutinized. In one pair of models, product modularity and process modularity are separate direct antecedents to manufacturing agility, which is modeled to affect firm growth performance; in a second pair of models, product and process modularity are related antecedents to manufacturing agility, with product modularity preceding process modularity. Results from the best-fitting model show that product modularity directly and positively affects process modularity, manufacturing agility, and firm growth performance. Process modularity was unrelated to manufacturing agility, and neither process modularity nor manufacturing agility predicted growth performance. Consistent with GMST, the study provides empirical evidence of the power of one element of a modular system to orchestrate a fit between a firm's product and manufacturing strategies and to directly drive system performance. Thus, modularity in product design is revealed as the key to understanding GMST effects concerning how changes in one system generate changes in other systems.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 184 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.