Between ‘Lean’ and ‘Reflective’ Production: The Case of Toyota Motors Thailand
Thunyalak Weerasombat and Ian Hampson*
Thammasat University,
Bangkok, Thailand
Australian School of Business,
University of New South Wales, Sydney, Australia
Prepared for the International Labour Process Conference, 27-29 March, 2012, Stockholm
Abstract
The Toyota Production System (TPS), popularly known as ‘lean production’, has been a controversial concept, at one stage hailed as the inevitable ‘best practice’ future of work. Against this view, variations of ‘lean production’ were noted in different national contexts, reflecting their shaping by different institutional environments. This posed the problem of identifying when any real production system was an instance of ‘lean production’ – in other words identifying the scope of the concept. Relatedly, there have been attempts to register correspondence between institutions and production systems, while critics of lean production have noted negative effects on workers linked to certain institutional configurations. Effectively acknowledging this, Toyota (among other Japanese auto manufacturers) adjusted its production system in certain plants, giving rise to a version of the Toyota Production System that some commentators called ‘post-lean’ because of its new emphasis on the Quality of Work Life (QWL), influenced by ‘Swedish’ production concepts that some have called ‘reflective’ production. This article addresses such questions through an investigation of the characteristics of the TPS implemented at Toyota Motor Thailand (TMT), the second largest production base of Toyota in Asia. It aims to locate the TPS between two ‘ideal types’ of production systems – ‘lean’ and ‘reflective’ production. Based on a wide range interviews and plant visits, this article argues that the form of TPS at TMT is best described as ‘lean’, although not fully so, as it does contain some ‘post-lean’ elements.
Keywords: Toyota; Toyota Production System (TPS); Lean Production, Reflective Production; Quality of Work Life (QWL); Toyota Motor Thailand (TMT)
* Contact author
Between ‘Lean’ and ‘Reflective’ Production: The case of Toyota Motor Thailand
Introduction
The Toyota Production System (TPS), most often known as ‘lean production’, has been a controversial concept, and there is no consensual ‘checklist’ of what makes a particular production system ‘lean’. As New argues[1], this makes it difficult to identify instances of ‘lean production’. The problem is significant if one wants to register national variations in production systems. Since our research problem was to trace the implementation of the TPS in Thailand, this required having some point of reference to say in what way it was different (or not) to other contexts. The first object of this paper is to propose a conceptual framework to help this identification.
An influential point of view associates ‘lean production’ with negative effects on workers, but Toyota itself, in an interesting period of experimentation in the 1990s in certain plants, adjusted its production system to emphasise Quality of Work Life (QWL). This was not as radical a change as may be imagined, since certain ‘forgotten’ TPS production concepts emphasize the importance of treating human resources with care – although the extent to which such prescriptions are followed varies with different societal and market conditions.[2] The resulting ‘version’ of the TPS some have labeled ‘post-lean’. This paper addresses such questions in the context of an empirical investigation of the characteristics of the TPS implemented at Toyota Motor Thailand (TMT), the third largest production base of Toyota in Asia. There are few studies of the TPS in Asia, and none that we know of in Thailand.
The paper locates the TPS at TMT on a conceptual framework we devised for the purpose, between two ‘ideal types’ of production systems – ‘lean’ and ‘reflective’ production. It argues that the form of TPS at TMT falls in between ‘lean’ and ‘post-lean’ versions. Rejecting questionable convergence models, a related stream of literature insists that the form taken by the TPS in any particular national institutional environment reflects that environment’s ‘societal effects’, along with ‘system’ and ‘dominance’ effects. In some respects the institutional environment in Thailand appears conducive to harsh, ‘lean’ versions of the TPS. Yet, perhaps surprisingly, some of the above mentioned ‘post lean’ characteristics are identifiable at TMT.
The paper first addresses the general issues, reviewing the characteristics of ‘lean production’ or the TPS, and introduces the ‘post lean’ perspective. We then explore the ‘institutional shaping’ of the TPS. Our argument is not ‘institutionally deterministic’, as we identify certain characteristics of the TPS at TMT that derive from Toyota’s strategic HRM in section three. Our fourth section proposes a conceptual model composed of several dimensions along which a production system could be located. Then we locate the TPS at TMS ‘between Lean and Reflective’ production in section five.
1. ‘Lean Production’ and the Toyota Production System
The first academic work discussing the TPS was by Sugimori et al who identified two basic concepts of the TPS: ‘cost reduction’ and ‘full utilisation of worker’s capabilities’.[3] Krafcik then coined the term ‘lean production’ to highlight the TPS’ emphasis on waste elimination.[4] Later, this concept was celebrated in influential book, ‘The Machine that Changed the World’ by Womack et al. These writers claim that ‘lean production’ is ‘lean’ because it uses less of everything, even as little as half of everything,[5] and that lean production is the most efficient and humane production system, which would replace the old paradigms of craft and mass production.[6] However, the claim that lean production is humane for workers was strongly opposed by a number of scholars, and the claims that ‘leanness’ contributes to efficiency were suspect because the book did not offer a sufficiently specific definition of ‘lean production’. As argued by New, the lack of clear criteria for ‘lean’ production ‘immunises’ it from ‘falsification’, because diverse practices can still be claimed to be lean production.[7] This is important when something (like productivity, efficiency, or work degradation) is seen to be an ‘effect’ of ‘lean production’, or if one or another component is missing from a particular implementation. In short, when is lean production really lean?
Accordingly, many attempts were made to capture lean production’s ‘true essence’. Studies from operational and management literature offer diverse sets of indicators to measure lean implementation. Karlsson & Ahilstrom assess lean production though eight items: waste elimination, continuous improvement, zero defect, JIT, pull instead of push, multifunctional teams, decentralised responsibility, vertical information system.[8] Meanwhile, Sanchez & Perez offered six groups of indicators: elimination of zero-value activities, continuous improvement, multifunctional teams, JIT production and delivery, supplier integration, and flexible information system.[9]
Other approaches identify sets of specific management practices and ‘tools’, or view the TPS as a philosophy.[10] Ohno himself stressed the ‘philosophy’ of ‘waste elimination’ in converting inputs to outputs, distinguishing valued-added from non-value-added steps. Kaizen (‘continuous improvement’) is a central feature.[11] Kaizen takes place through workers’ participation in contributing ideas to improve work processes. ‘Improvement’ means removal of all ‘waste’ (muda) in all activities of the production processes, including excessive inventory, unnecessary motions, overproduction, and idle time. It is driven by JIT, or producing and delivering only what is needed when it is needed.[12] This aims to ensure continuous flow in the production process. The tool that facilitates JIT is kanban, the printed cards identifying key information of the amount of items being produced and delivered to ensure minimised buffer/inventory in each production process.[13] Removing surplus buffers drives kaizen because it makes production imbalances or production problems easily noticeable, thus quick responses can be taken to maintain the continuous flow, and to improve it.
Muda is a generic term for ‘waste’, and includes inventory and, above all, idle time. Toyota texts also define two other aspects or causes of waste: muri and mura. The former is overburdened work imposed on workers/machines (overload), while the latter refers to inconsistent use of workers/machine (unevenness). Muri can lead to production defects and safety problems due to overloaded work; meanwhile mura causes the cycle of over-work and under-work, which is inefficient for the whole working time.[14] There is tension between these forms of ‘waste’, as, for example, eliminating the muda of idle time, can add to the muri of overloaded work. Problematising muri as waste indicates some potential for ‘humanisation’ of the TPS embedded in its core production concepts, depending on which are emphasised at particular times.[15]
Jidoka is a system to prevent defects that might block the flow of the production.[16] It utilises both mechanical-jidoka (equipment to detect abnormality and to stop automatically, e.g. poka-yoke or ‘fool-proof system’) and human-jidoka (workers note anything suspicious in the production flow, and can pull the so-called ‘andon cord’, to stop production).[17] This gives the team leader a chance to resolve the problem before the line stops. For the efficiency of jidoka, the plant layout is designed under the concept of ‘high visualisation’, making all processes easily visible.[18] Some studies also pinpoint the issue of the relation between waste elimination and heijunka or ‘levelled production’[19] to achieve continuous flow – thus eliminating waste from variation in production intensity (mura).
Critics argue that the TPS entails work intensification, causing high stress and risks to workers’ health.[20] Some scholars even renamed the TPS as ‘management by stress’.[21] For these researchers, superior organisational performance comes at the expense of workers’ well-being.[22]
On this view JIT pressures workers to work faster to produce smaller batches with shorter set-up times, while buffer minimisation increases dependency between work stations, reducing workers’ discretion over the pace of work, leaving very little ‘idle’ time for ‘personal breaks’.[23] Meanwhile, kaizen requires participation from workers, forcing them to often come up with more efficient ways to work, imposing tougher and tougher standards. [24] Standardised work fragments tasks, requiring workers to strictly follow preset work instructions, which results in the loss of work autonomy.[25] Team-based work may also lead to stress due to peer pressure. When a team member is absent or late to work, the rest of the team has to work harder to make up the shortfall or even to do OT to meet targets.[26] Such practices may not only cause stressful interpersonal relations but also reduced job satisfaction.[27] The mixed production of heijunka can lead to safety problem, as it requires workers to work on unfamiliar products.[28]
Yet Toyota literature argues that the TPS entails ‘respect for humanity’.[29] TPS advocates claimed that, under kaizen, workers have autonomy to suggest and change the work processes; meanwhile, the ‘andon cord’ and ‘jidoka’ make workers masters of the production processes as they have the authority to stop the production line if something goes wrong. Sugimori et al also argue that workers’ dignity is respected under the TPS because the system makes ‘full utilisation of workers’ capacities’ and entrusts them to exercise control over the work process.[30] Adler & Borys use the terms ‘enabling bureaucracy’[31] and ‘democratic Taylorism’[32] to argue that the TPS is human-oriented since workers are allowed and encouraged to design and control their work.
However such claims exist in some tension with received conceptualisations of the quality of work life (QWL). According to the Job Characteristic Model (JCM), the TPS is not human-oriented because work is highly fragmented, thus leading to low ‘identity’ and ‘significance’ of work, while repetitive and highly specified tasks deny workers ‘autonomy’.[33] This characteristic of work organisation may be a fundamental limitation on the extent of work humanisation that can be claimed to be associated with any particular version of the TPS.
Partly as a response to these criticisms, and partly as a response to political economic pressures deriving from a tight labour market, Toyota decided to experiment with ‘humanising’ its production system to improve the quality of work. In the 1990s, scholars reported a number of modifications in this regard and some even argued that a new ‘post-lean’ version of the TPS had emerged. This ‘post-lean’ version was said to be inspired by the general innovations in work humanisation that are associated with Scandinavian work design, the topic to be discussed in the following section.
‘Post-Lean’, and ‘Reflective’ production
The concept of ‘reflective’ production has become a banner for a ‘movement’ of alternative (to lean) traditions of work design .[34] It can be viewed as the opposite to lean production due to its emphasis on ‘human-centred’ work. Its origins lie in the famous production experiments of Volvo in the mid 1970s at the Kalmar vehicle assembly factory, followed up by those in the Uddevalla factory in the mid-1980s.[35] Facing demand increases in the 1980s, as well as difficulties in recruiting workers due to the image of tough and monotonous assembly work, Volvo reorganised work to be more attractive to workers.[36] A key principle was parallel, not linear, assembly flow and long-cycle work, bolstered by ideologies of worker professionalization supported by work autonomy, and independent assembly teams.[37] The innovations at Kalmar and Uddevalla were portrayed by many scholars as a production model associated with productivity, quality, and human-orientation, arguably the assembly design of the future.[38] Though these two plants were later shut down during the early 1990s, the outcomes of Volvo’s efforts to humanise production live on as exemplars. Also, they have inspired many modifications found in Toyota in the early 1990s.[39]
During the late 1980s, crisis in the Japanese labour market due to the bubble economy, labour shortages, and increasing power of the labour movement which problematised ‘work conditions’ – especially the phenomenon karoshi (death from overwork), pressured Toyota to make its work more attractive as it encountered difficulty in recruiting and retaining workers, especially the young, who refused to work in the automobile industry.[40] In 1992, the company introduced significant changes in the production line first at a new Miyata Plant in Kyushu and later to other plants. The changes included the segmentation of the assembly line and, in a major contradiction to ‘lean tradition’, the introduction of buffers to reduce interdependence among production lines. Workers were assigned more functionally-related (as opposed to unrelated, and therefore meaningless) tasks. Automation emphasised more ergonomic concerns.[41] All such changes focused on QWL improvement, aiming to make the TPS more attractive and friendlier to workers. Accordingly, some scholars argue that a humanised version of lean production, the ‘post-lean’ regime, had emerged.[42]