Introducing robust design in product development: Learning from an initiative at Volvo

Abstract

Robust design (RD) has a large potential to contribute to product and process improvements providing increased customer value. However, it has shown to be difficult to obtain these benefits in practice. This study aims to evaluate and learn from an initial approach to introducing RD within the Volvo Group. It is based on three pilot cases within the product development organisation of a business unit. Data were collected through formal interviews and informal dialogues with pilot participants, supplemented by existing documentation of the pilot cases. The main finding was that a RD initiative, characterised by ‘tool-pushing’ and with a predefined solution introduced by an external consultant, faced many obstacles and could not create a sustainable result. Instead, it was found that there is a need to involve engineers and create a learning culture in which RD principles can become a natural part of work practices. This study identified six obstacles to the success of the initiative, which were perceived as learning points for a broader application of RD at the company. This underscores that RD initiatives can also be hampered by similar types of obstacles that have been identified in research on other change processes.

Keywords:

• change management
• resistance to change
• localising
• middle management
• product development
• RD method
• P-diagram
• Pugh matrix

Notes

The Volvo Group includes companies producing trucks, buses, construction equipment and marine and airplane engines. Volvo was founded in 1927.

This is expressed in Taguchi's ‘quality loss function’ that states that any deviation from the target value leads to a quadratic loss in quality (or customer satisfaction).

RDM is referred to as any systematic effort (statistical or non-statistical) to achieve insensitivity to noise factors (the sources of variation) (Arvidsson & Gremyr, 2008).

Moosa and Sajid (2010, p. 754) commented that ‘Even the most scholarly literature available is focused on principles, phenomena, and techniques of Six Sigma. Consequently, social aspects of implementation are underestimated, undermined or undiscovered during the implementation phase’.

Dahlgaard-Park (2011) observed that both organisations and many theoreticians focus on mechanistic and rational aspects of organisations, including believing ‘that measurements can cover every aspect of organisational improvement’ instead of ‘the more “softer” aspects such as organisational culture, values, people motivation, training & education, etc.’.

Individual managers from Volvo also participated in an external one-day training programme with Shin Taguchi called ‘Introduction to Robust Design using Taguchi Methods’.

The RD method, as referred to in this article, mainly concerns Taguchi's methods for system, parameter, and tolerance design.

These tools included the Pugh concept evaluation and selection of concepts (system design step), the P-diagram for functional analysis, and DoE (parameter and tolerance design steps).

The P-diagram provides a description of a design concept in terms of the signal (input) and response (output) variables, the factors beyond the control of the designer (noise factors) and the factors that can be specified and set (control factors) in order to minimise the noise factors' effect on the response variable, that is, RD.

The four standard categories of the projects' deliverables in the company.

It should be stressed that this is a belief that top management themselves know the project's scope and priorities – which, however, is not proven.

This is an area where one of the practices identified by Hasenkamp et al. (2009) could be helpful, ‘Focus on the customer’. This practice concerns a systematic way of identifying key customer characteristics and translating them into quality characteristics that RDM aims at making insensitive to noise factors.

Dahlgaard, Pettersen, and Dahlgaard-Park (2011) emphasise the importance of involving people in identifying problems, root causes and needed actions as when change is more profound, ‘It requires that people understand what to change and why’.

Hasenkamp et al. (2007) provide guidelines for when RD tools potentially could be useful.

Alänge (1992, p. 162) pointed out ‘that the role of first line supervisors and middle management is crucial to the survival of the QC concepts’.