[DSRP Evidence](https://dsrpevidence.org/)

# The role of product architecture in the manufacturing firm

## Details

**Authors** Ulrich, Karl

**Year** 1995

**Publisher** Research Policy

**Kind of work** article

**Discipline** Business & Management

**Secondary disciplines** Engineering

**Applied** false

[Read it at the publisher](https://doi.org/10.1016/0048-7333(94)00775-3) 
10.1016/0048-7333(94)00775-3

## In authors' words

### Abstract

Product architecture is the scheme by which the function of a product is allocated to physical components. This paper further defines product architecture, provides a typology of product architectures, and articulates the potential linkages between the architecture of the product and five areas of managerial importance: (1) product change; (2) product variety; (3) component standardization; (4) product performance; and (5) product development management. The paper is conceptual and foundational, synthesizing fragments from several different disciplines, including software engineering, design theory, operations management and product development management. The paper is intended to raise awareness of the far-reaching implications of the architecture of the product, to create a vocabulary for discussing and addressing the decisions and issues that are linked to product architecture, and to identify and discuss specific trade-offs associated with the choice of a product architecture.

### What they set out to do (purpose)

To define product architecture more precisely, provide a typology of product architectures, and set out the linkages between a product's architecture and five areas of managerial importance: product change, product variety, component standardization, product performance, and product development management.

### Who or what was studied (sample)

No empirical sample. The argument is developed through worked examples, chiefly a utility trailer, together with desks, personal computers, the Sony Walkman, motorcycles and automobiles.

### How they did it (methods)

Conceptual synthesis of design theory, software engineering, operations management and product development management, illustrated with comparative design examples, including a count of the component types needed to offer 108 trailer variants under a modular and an integral architecture.

### What they found (results)

Product architecture is defined as (1) the arrangement of functional elements, (2) the mapping from functional elements to physical components, and (3) the specification of the interfaces among interacting components. A modular architecture has a one-to-one mapping from functional elements to components and de-coupled interfaces, with three subtypes (slot, bus, sectional); an integral architecture has a complex mapping and/or coupled interfaces. Modular architectures localize change to single components, allow variety to be built combinatorially (108 trailer variants from 12 component types, against 73 types for the integral design), make component standardization possible, and favor local performance characteristics. Global performance characteristics such as mass, size and drag can only be optimized through integral architectures, via function sharing and geometric nesting. No single architecture is optimal in all cases; most products are hybrids, and the choice of architecture shapes how product development is organized and managed.

## Commentary

### In short

Ulrich defines a product's architecture as the way its functions are divided among physical parts and the way those parts connect. He shows that this one design choice governs how easily a manufacturer can change, vary, standardize and improve what it makes.

### In more detail

Ulrich sets out to give engineers and managers a shared vocabulary for a decision that is usually made implicitly. He defines a product's architecture by three things: how its functions are arranged, how those functions are assigned to physical components, and how the components meet at their interfaces. Using a utility trailer as a running example, he contrasts a modular design, in which each function lives in exactly one component and a change to one component does not force changes to its neighbors, with an integral design, in which functions are spread across components and the components are tightly interdependent. He divides modular designs into slot, bus and sectional types according to how the components connect. He then traces the consequences across five areas. Modular designs let a firm upgrade or replace one function at a time, build wide product variety by recombining a small set of parts (12 component types yield 108 trailer variants, where the integral trailer needs 73), and use standard components from suppliers. Integral designs are needed when a product's most valued qualities, such as weight, size, drag or noise, depend on every part at once, because sharing functions across parts and nesting them tightly saves mass and space. The choice also reshapes the organization: modular products allow parallel, specialized, supplier-like development, while integral products require a core team that tunes the whole system together. Ulrich concludes that no architecture is best in general, that most products mix the two, and that a component for one firm is itself a whole product for its supplier.

### Response

Ulrich's definition lines up closely with three of the four patterns. Functional elements and physical components are parts of a product at two levels of description, and the mapping between them is a part-whole question (S). The interfaces, and the test of whether a change to one component forces a change to another, are relationships (R). The typology itself is built from distinctions: modular against integral, coupled against de-coupled, local against global performance (D). Ulrich also notes that a product can appear modular or integral depending on the level at which it is observed, and that a component to one firm is a system to its supplier; this points toward perspective, but it is a claim about level of decomposition more than about an observer's vantage point, so P is not claimed. The caveats are significant. The paper is conceptual, and its claims about effects on firm performance rest on worked examples and argument rather than measurement. It also draws directly on the modularity literature of software engineering, so it is an elaboration within design theory rather than a wholly separate line of discovery.

**Patterns it shows** D, S, R

Shows it makes a difference

**Added** 2026-11-20

**How to cite this** Ulrich, Karl (1995). The role of product architecture in the manufacturing firm. Research Policy.
