Sociotechnical systems: towards an organizational learning approach

J. Eng. Technol. Manage. 18 (2001) 271–294

Sociotechnical systems: towards an organizational learning approach Eric Molleman∗ , Manda Broekhuis Faculty of Management and Organization, University of Groningen, P.O. Box 800, 9700 AV Groningen, The Netherlands

Abstract By means of three design principles (the sociotechnical criterion, the principle of minimal critical specification and the principle of joint optimization of the technical and social system), STS as a design theory is related to four organizational performance indicators (price, quality, flexibility and innovation). As a diagnostic theory, STS helps to find contingencies between environmental demands and work design. The diagnoses result in sets of STS practices. It is argued that as long as price and quality are the only important performance criteria, STS practices have little to offer and their contributions will be only at the job level. If flexibility is of importance, STS has much more to offer, on the job level as well as the organizational level. The same is true for when innovation is a relevant indicator, in which case STS practices may also help to ‘design’ processes, such as mutual trust among workers and diversity with respect to attitudes, abilities and cognitions. It is argued that the dominant performance indicators have changed in a cumulative way from efficiency, via quality and flexibility towards innovation and learning. In accordance with these changes, the STS principles are extended with the concept of organizational learning. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Sociotechnical systems; Performance indicators; Organizational learning

1. Introduction In this paper, we will present the sociotechnical systems (STS) theory as a diagnostic theory and as a design theory. We define STS as an integral theory of work design and quality of working life (QWL). By means of three principles of STS, we will diagnose what kind of work design may help organizations achieve four different patterns of performance indicators. In other words, with a specific pattern of performance indicators in mind, we will depict a work design contingent on these three principles. Considering STS as a practical ∗ Corresponding author. Tel.: +31-50363-3846; fax: +31-50363-2032. E-mail address: [email protected] (E. Molleman).

0923-4748/01/$ – see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 3 - 4 7 4 8 ( 0 1 ) 0 0 0 3 8 - 8

272

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

design theory, we will examine whether STS practices, such as job enrichment, job enlargement and self-managing teams, fit in the diagnosis or whether non-STS practices seem to be more appropriate. Our primary aim is to contribute to the development of STS theory and more specifically to extend it to the phenomenon of organizational learning. Moreover, the links we make between performance patterns and the success of STS practices may aid the understanding of previous conflicting research outcomes. In this section, we will systematically introduce our framework. First, we will describe the three diagnostic principles and following this we will consider four patterns of performance indicators. Since the early years of STS theory, many different organizational principles have been launched, which we may find, amongst others, in the works of Emery (1969, 1978), Herbst (1974), Cherns (1987), Pasmore (1988), and Pava (1986). We will determine the value of STS in terms of dealing with different performance indicators by means of three of these principles in particular, the sociotechnical criterion, the principle of minimal critical specification and the principle of the joint optimization of the technical and the social system. The sociotechnical criterion deals with the control of variance and states that variances should be controlled as near to their point of origin as possible (Cherns, 1987). The sociotechnical criterion was incorporated in STS from systems theory, where it was referred to as ‘the principle of requisite variety’ (Ashby, 1969). According to this principle, to manage environmental demands successfully, an organization should have enough means to transform the input of information, materials and parts into the output that it desires, that is, only variety can beat variety. The principle of minimal critical specification refers to the following: define as little as possible how a worker should perform tasks, but provide just enough directives to ensure that he or she is able to perform the task properly while still allowing for the employee’s personal contribution (Cherns, 1987; Morgan, 1986). This refers particularly to local autonomy and decentralized control, which will result in enriched jobs and empowered workers. The joint optimization principle deals with the fact that STS endeavors to consider both the social and the technical system simultaneously. The technical system refers to the production structure, the technical equipment and to systems from the field of information and communication technology. The social system refers to human resources, job design and to the control structure. We will discuss the relevance of STS in the light of four performance indicators: price, quality, flexibility and innovation. Kumpe and Bolwijn (1994) have discussed these performance indicators and have placed them in a historical perspective. They argue that until the 1960s, price was the only leading objective and that in the 1970s, quality also became an important indicator. In the 1980s, the need for flexibility grew and in this day and age, they consider innovation to be the major value-added criterion. Nowadays, more and more organizations have to deal with highly dynamic environments and complex and dramatically changing transformation processes, making flexibility and innovation key issues for most firms (Volberda, 1996). Kumpe and Bolwijn consider these performance indicators to be cumulative, that is, first only price was the leading indicator, then price and quality, and so on. We will follow this cumulative approach and we will not detract from their historical perspective, although we think that not all of the four performance indicators are significant

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

273

for every organization nowadays and that, for example, we may still find firms whose only leading competitive strength is price. In the first of the four sections that follow Section 2, we will focus on firms which operate in markets which predominantly demand low-priced goods, making efficiency the main organizational issue. In Section 3, we will discuss markets on which quality is also an important demand characteristic. In Section 4, we will look at what STS has to offer when firms seek (certain types of) flexibility. In Section 5, we will discuss organizations which face demands for innovation and customization, making the ‘learning organization’ the main issue. Besides the performance indicators above, in each section we will consider the QWL, which may be considered to be the original leading STS objective (e.g. Emery, 1969; Trist and Murray, 1993; Pasmore, 1995). Following the argument that flexibility and innovation are key indicators nowadays and therefore deserve most intention, we will move more quickly through the sections on ‘price’ and ‘quality’ than through those concerning ‘flexibility’ and ‘innovation’. Within each section, the diagnoses pertaining to the usefulness of STS with respect to attaining specific performance indicators will be followed by STS design practices. More specifically, we will consider how the three principles may be helpful in designing jobs and organizational structures that will facilitate the realization of these performance indicators. We realize that in engineering and technology management there are many useful non-STS related manufacturing practices that have been developed to cope with specific environmental demands. Our focus, however, will be on the way in which STS practices may help to deal with each of these demands, and we will refer to some of the prominent coping strategies, especially when our diagnosis indicates that STS practices have little or insufficient to offer. 2. The hard fit between STS and price 2.1. Diagnosis Firms which operate in markets where price is the only significant performance indicator can predominantly be typified as mass-producing firms. Such firms make large quantities of a very limited set of products and this limited variety in demand causes a restricted diversity in work processes. In such a situation, standardization of work processes is not only relatively easy to attain but also has a strong competitive value. It contributes to efficiency, given the fact that it is less costly than other forms of coordination (cf. Thompson, 1967). It is more efficacious at a local level to limit repeated discussions about the best way to do things and to establish what the best practices are and have everybody adhere to them. The necessity for local autonomy and decision-making appears to be minimal. In fact, according to the sociotechnical criterion, the need for local autonomy and decision-making concerning these processes is minimal, which makes a work setting with maximum specification more likely than one with minimum specification. With respect to the joint optimization principle, we assume that the technical system will completely dominate the social system. Mechanization and automation will aim at speeding up production and the pace of workers, as well as making workers redundant (Benders, 1993;

274

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

Wall et al., 1987; Mintzberg and Quinn, 1992). As technology evolves, it becomes more possible in these production environments to substitute equipment for human labor, and thus to reduce the coordination that humans necessarily bring to their tasks. The technical systems help to standardize work processes and their use is principally directed towards the necessity of efficiency and towards the substitution of labor. The new assembly lines which Japanese car makers have recently been opening illustrates this. For example, at Mazda’s Hofu plant, automation in final assembly started at 18% and aims to reach 50% (Williams et al., 1992). 2.2. Practices The diagnosis depicted above suggests that STS practices will have no chance in a mass-producing firm, because there is no direct positive relation between the implementation of an STS design and performance, as long as efficiency is the leading performance indicator. However, some practices may indirectly contribute to efficiency. First, the classic study of Trist and Bamforth (1951) in the English coal mines has shown that allowing a group of workers to make decisions themselves about, for example, the planning and the assignment of tasks to team members affects the social relations among workers in a positive way, enhancing their motivation. Second, even if the need for local decision-making is low, job enlargement will contribute to task identity, which increases the meaningfulness of the job (Hackman and Oldham, 1980). Combining tasks enlarges the cycle time and provides a sense of completeness, which is an important element in improving the work content. The integration of tasks reduces their repetitiveness and it is likely that this will improve the QWL and the motivation of workers. A company with a highly motivated workforce may easily outperform one involving work designs with only repetitive routinized and short cyclical tasks. However, Adler and Cole (1993) concluded in their study that simply bolting and screwing together a large number of parts in a long cycle does not change the real problem, which is the separation of ‘thinking’ and ‘doing’. We are inclined to conclude that in cases where efficiency is the only leading performance indicator, STS practices will only marginally be able to contribute to this (see also, Manz, 1992; Dunphy and Bryant, 1996). With respect to non-STS design principles, we think that management will adhere to practices favored by, for example, lean production (Womack et al., 1990; Niepce and Molleman, 1996). Lean production tries to achieve a perfectly balanced production system with the absence of buffers. Through meticulous time-and-motion studies exact standards are developed for each process. These specifications are extremely detailed. Instead of having the freedom to work when they wish, workers have to adhere to a fixed pace, as the reduction of inventory buffers makes workers increasingly dependent on work-flow time-sequencing governed by the technology employed. All wasteful motions in the performance of the job are eliminated, which leads to the creation of narrowly defined and simple jobs. Such a system that gets rid of any buffers, makes every disturbance instantly visible. The practice at Toyota, for example, is to constantly withdraw resources from the line to see what happens; the point where a disorder emerges provides the focus for further increasing efficiency. The aim is to discover the minimum necessary to achieve the job properly first time round (Schonberger, 1982; Krafcik, 1988). It is obvious that, as far as job content is concerned, the work system depicted above will lead to a poor QWL (e.g. Turnbull, 1988).

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

275

3. Quality and price: specification and standardization 3.1. Diagnosis When quality in the sense of conforming to requirements (Crosby, 1979) is the main performance indicator besides efficiency, it becomes important to control undesired variations in quality where these emerge, to minimize the number of defective goods. When these variations arise during manufacturing, it is important, according to the sociotechnical criterion, to delegate these quality control tasks to the shop floor (see, for example, Hut and Molleman, 1998; Balkema and Molleman, 1999). However, if quality and efficiency are the only dominant performance indicators, this still means predominantly routinized work in which standardized tasks have to be done, the best way in which to do them is known and specified, and the required output has been precisely defined. If the output deviates from its standard, the actor is supposed to receive feedback indicating that he or she has deviated from the standardized work process. The actor should correct his or her behavior and feedback is used to inform the worker of the extent to which his or her performance deviates from fixed criteria. In such environments, quality control is not much more than quality assurance in which the ultimate goal is the absence of process variability, zero defects, and complete reliability. Although quality control and the control of variability is part of the workers’ jobs, these tasks are still highly specified and local decision-making is very limited to standard procedures. Moreover, technical systems support these quality control objectives and dominate and form the workers’ jobs. These points lead to the conclusion that STS may contribute somewhat more to performance in the case of quality being a second indicator besides efficiency than when efficiency is the only significant one. 3.2. Practices Job enlargement in the form of the horizontal integration of tasks may contribute substantially to higher quality performance. As stated in the previous section, such an integration may increase job identity and cycle time and will reduce the repetitiveness of tasks. Moreover, it is likely that workers will gain better insight into the way the different processing steps influence each other. The workers will be able to observe and correct deviations at an earlier stage, which will improve quality. The contribution of job enrichment to higher levels of QWL is limited. Although, the assignment of control tasks to the shop floor workers seems to result in enriched jobs, we have to realize that these tasks are highly specified and that in fact enrichment is very confined. It is likely that management will adhere to non-STS practices, such as ‘continuous improvement’, which are offered by philosophies such as lean production or shop floor management and which in fact decouple ‘enrichment’ from daily work (Schonberger, 1982; Shingo, 1986; Suzaki, 1993). Kaizen, the principle of ‘continuous improvement’, is based on the premise that there is always a better way of doing things. Continuous improvement means the continual rationalization of production, since in its search for the minimization of waste, it aims at a reduction in activities and the minimization of quality problems. The concept is clearly structured and

276

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

disciplined as workers gather in so-called ‘quality circles’, separate from their daily work. On a central level, specialists and experts are required to judge the impact of a local improvement on other processing steps, before an organization changes a working method. Continuous improvement structures the involvement of workers and, in fact, standardizes improvement activities and creativity. Workers’ jobs are enriched and enlarged, insofar as they are indeed invited to find better ways to streamline the system further (Hitomi, 1993). In accordance with Argyris and Schön (1996), this refers to single-loop learning, that is, to processes which may induce improved actions leading to prescribed outcomes. Nevertheless, this is a far cry from what they refer to as a ‘learning organization’, a phenomenon we will deal with in a later section. Conti and Warner (1993) describe quality circles as a system in which employees spend four hours a month making their work for the rest of the month even more Taylor-like. The result is a high-stress system that only emphasizes the production side and does not meet the STS objective of QWL (Vogel, 1979; Parker and Slaughter, 1992). We conclude that when quality, as in confirming to specifications, and efficiency are the only main objectives, STS practices may, to a certain extent, help to reach these goals. Nevertheless, we think that practices that focus on specifying as many critical issues as possible will dominate. These practices will use technical systems to control processes and will limit the involvement of workers. For a systematic comparison of quality management and STS we refer to Manz and Stewart (1997), and for an analysis of lean production versus STS we refer to Niepce and Molleman (1998). 4. The answer of STS to flexibility: IOR 4.1. Diagnosis When consumers and (industrial) clients demand more diversity and shorter delivery times, while also being critical with respect to the price/quality ratio of products, flexibility becomes an extra performance indicator. This situation pertains to organizations which produce a lot of different products in many variants. The repetitiveness of orders is relatively low and the manufacturing process is relatively complex. With respect to the sociotechnical criterion, flexibility means that an organization has to deal with much more variety in environmental demand. According to the principle of minimal critical specification, it becomes apparent that to respond alertly and effectively to variety in customer demand, it is important for there to be a lot of leeway for local decision-making. An advanced STS approach developed in the 1980s, pleading for ‘complex jobs within simple structures’, assumes the ability to cope with flexibility properly as well as with efficiency and quality. With respect to the principle of joint optimization, it states that an integral redesign of technical and social systems will contribute to these performance indicators. This approach was developed in The Netherlands by a group of scientists and consultants around De Sitter (De Sitter et al., 1986; Kuipers and Van Amelsfoort, 1990). Their design practices aimed at more efficiency, more transparent control structures, lower coordination costs, higher levels of organizational flexibility and improved performance, such as lower throughput times and a better delivery performance, besides desirable jobs

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

277

and higher levels of QWL. Such objectives could be sold more easily to managers than just ‘soft’ issues, such as QWL. Their ideas, which they labeled modern sociotechnical theory (MST) or integral organizational renewal (IOR), emerged within a faculty of Industrial Engineering in which engineers, consultants and social scientists worked together, accidentally leading to their knowledge fields merging. The engineers contributed subjects, such as group technology and cellular manufacturing and the social scientists the old STS principles and related knowledge. This integration not only resulted in a more advanced interpretation of the old STS principles, but also in an approach which received more recognition from engineers working in industrial settings. In our view, this facilitated its dissemination and implementation considerably. Unfortunately, up to the 1990s, their work did not appear in the international (English/US) literature or only scarcely (e.g. Van der Zwaan, 1975). In the 1990s, they started to discuss their organizational design principles more extensively in the international forums (e.g. Van Eijnatten, 1993; Dankbaar, 1997; De Sitter et al., 1997; Van Eijnatten and Van der Zwaan, 1998). In the last decades, hundreds of Dutch firms have successfully implemented their principles, either partly or entirely. We will now discuss the main design practices of their approach. 4.2. Practices The IOR starts with a strategic orientation of the firm, and then goes on to (re)design the production or ‘technical’ structure from the macro (firm) level down to the micro level (assigning resources to cells), with the purpose of ‘paralleling’ work processes or work flows. More specifically, the first step in designing the production structure concerns the gathering of product characteristics and process characteristics, such as batch sizes, processing steps, and routings by means of, for example, production flow analyses (Burbidge, 1992). Next, these data are related to machines and other resources by putting them into matrices and then these matrices are divided into sub-matrices (groups) by using, for example, clustering techniques (for an overview, see Suresh and Kay, 1998). Resources are assigned to cells or groups in such a way that cells, as far as possible, are able to do all the processing with respect to a limited range of products or clients. This leads to ‘paralleled’ flows. The number of goods or services that needs processing in several groups is minimized and, through this, the need for time-consuming and costly coordination between units decreases, making short delivery times more realistic (Thompson, 1967; Galbraith, 1973; Cummings and Blumberg, 1987). In line with the sociotechnical criterion, the potential responsiveness (flexibility) of groups towards changes and diversity in customer demand will be large, because responses do not affect other groups. These groups may be regarded as teams which are allowed to make autonomous decisions with respect to the transactions with customers and to the way the transformation processes have to be organized to achieve the intended output. When the technical structure has been completed in a top–down way, the work design and control structure (the ‘social’ structure) is designed bottom-up (from micro to macro level), in which a distinction is made between job enlargement and the control structure (job enrichment or empowerment). With respect to job enlargement, the focus is on raising the level of multi-functionality of team members and on the integration of jobs. Multi-functionality

278

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

creates a buffer of capacity so that a team can deal more effectively with shifts in demand, in this way increasing the flexibility of a team (e.g. Ebeling and Lee, 1994). Another important potential benefit of multi-functionality is the opportunity for the horizontal integration of tasks and the reduction of the division of labor, to which we already referred in a previous section. If related tasks are integrated, the identity of the job will increase, the need for coordination between workers will be reduced and the autonomy of the individual worker will be enhanced (Huber and Brown, 1991). This contributes to efficiency and quality control as well as to QWL. The control structure is designed according to the principle of minimal specification: create as much local autonomy and decentralized control as possible. With respect to the control structure, the premise is that all control tasks are assigned to the lowest organizational level (the group) and only if there are sound arguments not to allocate them there are they assigned to a higher level. The result is that most of the control tasks of (lower) managers are moved to the team. The managers’ jobs are redesigned by the integration of functional management areas. As has been advocated in human resource management literature (e.g. Legge, 1991), to increase local responsiveness (flexibility) to customer demands, it is necessary to strengthen the autonomy of line management by integrating areas such as personnel, production and financial management. This approach not only stresses the importance of the autonomy of production cells, but also the fact that supporting staff should function under the supervision of a team leader. Although many firms have adopted the IOR principles, in practice there have proved to be various reasons why a full implementation has often not been accomplished (e.g. Suresh, 1992; Fazakerley, 1976; Molleman and Van Knippenberg, 1995; Balkema and Molleman, 1999). We will just refer to one reason. The possibility of creating full autonomous cells without inter-group relations may be limited, as it depends, among other things, on the divisibility of resources. For example, machines may be unique, which means that their processing can not be transferred to other machines. These unique machines are often the most expensive and (therefore) bottleneck devices. Similar arguments may be valid for workers with unique qualifications. Other reasons for the indivisibility of equipment may be their processing characteristics (galvanic plating, electrostatic painting) or security reasons, as in the case of radiation devices in hospitals, for example. Moreover, if equipment is very expensive, such as flexible manufacturing systems or a nuclear magnetic resonance spectroscope, management will try to maximize the utilization level of these means and adjust other processes to these ‘functions’. In this section, we have argued that some markets demand flexibility in terms of diversity and delivery times, as well as quality and efficiency. The IOR approach helps to achieve these objectives as well as high QWL. The IOR practices follow the sociotechnical criterion and the principle of minimal critical specification by creating units which are able to deal autonomously with local variances. The technical and the social system are not really designed jointly, since the technical (production) system is designed first, followed by the social system (job design and control structure). Nevertheless, the attention paid to both systems is systematic and considerable in comparison to non-STS practices aimed at flexibility. Lean production, for example, strives towards a total absence of buffers by using ‘just-intime’ practices, which are directed towards producing variants of products without creating

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

279

buffers, large amounts of work-in-progress or stocks of end products. Advanced technical systems are used to cope with the increasing amount of variety but in essence, these practices still aim at standardization and the division of labor, and at simple tasks which workers can master in a short time to make them flexible (Delbridge et al., 1992). Single minute exchange of die (Shingo, 1985) is another practice to enhance flexibility. It aims at the reduction of set-up time and the maximization of production time of machines or technical systems in general. A third example is the design of products in a modular way. Up to a certain point in the manufacturing process, a firm may produce large numbers of standardized semimanufactured articles for stock, decoupled from specific customer orders. After this point, however, the firm can transform or combine these articles according to specific customer wishes into numerous kinds of end products and hence be highly responsive to variety in customer demand. Nevertheless, these practices pay no attention to the social system and have no intention of improving the QWL. We tentatively conclude that the way in which the IOR approach employs the STS principles contributes substantially to the realization of the above-mentioned performance indicators. 5. Innovation and creation of knowledge: a learning approach to STS 5.1. Diagnosis Besides the performance criteria mentioned in the previous sections, innovation seems to have become a major determinant for long-term success, for organizations that operate in highly dynamic markets. Such organizations often have (at least partly) a proactive strategy, that is, they do not deal with environmental demands in a reactive way only, but they also shape environmental demands. Innovative firms create orders and products which are unique. This causes processing that is non-routine and non-repetitive in nature. The course and outcome of each processing step may vary considerably and the outcome of each action will also affect the content and course of other processing steps. Variation has its point of origin in the processing itself. Applying the sociotechnical criterion, which states that variation should be controlled as closely as possible to its point of origin, creates a need for actors who are skillful in problem-solving, in coping with ambiguity, in decision-making concerning the objectives to strive after and concerning the way to reach these while doing the job (see, for example, Cummings and Blumberg, 1987; Manz, 1992). In terms of the principle of minimal critical specification, it is directly (inversely) related to the extent to which an outcome and the processes leading to this outcome can be specified beforehand. If the output is not exactly prescribed, the cycle of ‘analyses-diagnoses-planaction-evaluation’ has to be reconsidered continually by re-appraising problems, norms, working methods and desired outcomes (Kolb, 1984), a cycle which Argyris and Schön (1996) have referred to as ‘double-loop learning’. Organizations that opt for innovation will create a competitive advantage by coming up with new ideas and by finding solutions for unique and complex problems. This makes it clear that the workers, the social system, will be the most critical production factor, while the technical system will be employed to support and optimize the use of human competences (Schneider, 1994).

280

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

The realization of unique and complex products mostly requires such a diversity of knowledge and abilities that it is not realistic to expect individual workers to have all this knowledge and to master all relevant abilities. Consequently, if an order has been assigned to a group or team of workers, these workers will be highly interdependent because, as stated before, the course and outcome of each action will greatly affect others. According to Thompson (1967), workers will be mutually interdependent, which is the strongest type of interdependence. Cooperation and mutual adjustment particularly enhance performance in this case. Mutual adjustment implies that the creative solutions one member finds have impact on the work of others and vice versa: ‘what you do affects what I have to do and vice versa, so let us align our activities’. In fact, cooperation should go one step further than mutual adjustment. It should involve finding solutions together, learning from each other and finding synergy in innovative and creative processes. The above-mentioned loop of analyses-diagnoses-plan-action-evaluation and its reconsideration, are at least partly, collective activities. These collective activities may contribute substantially to the creation of knowledge and shared meanings, which helps to keep the firm innovative. Therefore, the creation of knowledge and shared meanings will become another (related) performance indicator for the organization itself. This means that creative and learning processes can be as important as the processing of the products and services themselves. Learning through meetings, interpersonal communications, reading or off-the-job training is not a new phenomenon. In the above-mentioned context, however, it is part of the primary process of organizations as well as an objective thereof. Learning refers to a process in which collective meanings are developed (Cook and Yanow, 1993); it is an ongoing problem-oriented process which is tightly connected to daily work. In this section, we will consider what the STS principles may offer to stimulate learning processes and to create a learning organization. We think that these principles should pertain to design and process issues, that is, practices leading to structures and processes that shape optimal conditions for learning and innovation. Although the design and process issues appear to be narrowly interwoven, we will briefly discuss some design issues first, and then focus on individual and interpersonal characteristics that affect innovation and learning processes. We believe that the STS principles can be elaborated to become a theory of organizational learning that is valuable to organizations whose leading competitive performance indicators are learning and innovation. However, to accomplish this, we will utilize the existing knowledge and models pertaining to the phenomenon of the learning organization and place STS in such a framework. Therefore, we will first use this knowledge to derive design and process practices that support a learning organization, for which we will refer to the work of Senge (1990). Next we will relate these practices to the STS principles that form the backbone of our paper. Senge distinguishes five factors (which he calls ‘disciplines’) that contribute to a learning organization. Since these are rather abstract, we have felt free to interpret them in a more operational way. The first factor pertains to ‘personal mastership’, which refers to skills such as reflective, research, problem-solving and social skills. We consider this factor more generally to refer to the characteristics of individual workers that are related to learning. We consider his second ‘discipline’, the use of mental models, to mediate between the first factor and ‘team learning’, the fourth ‘discipline’ in Senge’s framework. Individual

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

281

workers have their own personal past, experience and knowledge, leading to unique perceptions and interpretations laid down in mental models. These mental models are exchanged, discussed and altered during interactions with others. Senge considers his third ‘discipline’, the emergence of a shared vision, primarily to be a condition for organizational learning. We think that by placing the development of shared visions in a learning cycle, it may also be considered to be the outcome of collective learning at a strategic level, while being a condition for learning at a more operational level. Likewise, the shared meanings that are the outcome of these learning processes may be conditions for further learning. Therefore, we will regard Senge’s third ‘discipline’ as a local and temporary learning outcome of the aforementioned interaction processes. The fifth ‘discipline’, systems thinking, may help to relate these local and temporary meanings, which arise between individuals or within groups of workers, to learning at the organizational level. We will deal with this facet in Section 5.2 that may foster organizational learning. We will also relate them to the STS principles and make some propositions that proceed from our arguments. We assume these propositions are valid for organizations whose leading competitive performance indicators are learning and innovation. 5.2. Design practices If innovation and the creation of unique products or services are what is required, fixed teams with clear boundaries, which IOR opts for, may easily fail. The combination of skills and resources needed for a specific order may be so unique that a fixed grouping of means will not suffice. It will be necessary to group production means around individual orders and to form temporary (project) groups (see, for example, Francis and Winstanley, 1988). To be able to assign resources in a flexible way to orders or projects, and to facilitate learning processes, a matrix structure seems to be the right organizational design. Besides temporary project groups, organizations with matrix structures often have fixed structural ‘functional units’ in which employees with similar competences are grouped. Within these groups, obtaining and enlarging specialist knowledge is a central issue. Such a design can easily be related to the work of Daft and Huber (1987). They relate the way in which learning takes place within organizations to two dimensions: the equivocality of information and the amount of information. Equivocality has to do with the ambiguity of the problem to be solved. By combining these two dimensions they depict four modes of learning. The functional units are located in the quadrant of low equivocality and high amount of information. In these mono-disciplinary groups, we may find specialists and experts who share a lot of knowledge and who have mostly similar mental representations working on problems with, according to Senge, ‘detail complexity’. Temporary projects that, according to Senge, deal with complex and ambiguous problems are located in the quadrant with high equivocality and high amount of information. Project teams are mostly multi-disciplinary, composed of workers from different ‘functional groups’ and will, in terms of Senge’s first ‘discipline’, be more diverse with respect to learning styles, skills, knowledge, attitudes and experiences. Therefore, these projects are a good potential place to discuss the members’ mental models (second discipline) and to obtain and create knowledge that exceeds their own domain. Under the right conditions, which we

282

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

will discuss in Section 5.3, this will result in shared meanings within a team and in team learning (Senge’s third and fourth disciplines). When workers participate successively or simultaneously in different project teams, they are able to disseminate their new locally shared experiences, knowledge and mental models to other parts of the organization (Golembiewski, 1995). In this way, organizations may become social networks in which locally shared meanings are disseminated and contribute to learning at a higher level, that is, organizational learning. We have related this dissemination process to Senge’s fifth factor concerning ‘systems thinking’. If we relate the design practices depicted above to the sociotechnical criterion, we recall that a complex and dynamic environment demands ‘variety in design’, that is, flexible designs such as matrix structures. According to the principle of minimal critical specification, management has to specify and create good learning conditions, an issue we will deal with in Section 5.3. With respect to the principle of joint optimization, we wish to bring to the fore that nowadays, there are several methods and ‘technical systems’ in the area of information and communication technology, group support systems and multi-media applications that support project planning, group learning and the dissemination of team learning at the organizational level (see, for example, Hauser and Clausing, 1988; Jessup and Valacich, 1993; Muller, 1999). The practices presented in this sub-section lead to the following propositions. Proposition 1a. Creating flexible structures, such as a matrix structure with functional groups and project teams will promote team and organizational learning. Proposition 1b. Specific technical systems will support team and organizational learning. 5.3. Practices: characteristics of individual workers In this sub-section, we will focus on Senge’s first ‘discipline’, personal mastership, the degree to which individuals are able to contribute to creative and learning processes. We realize that we present only a fraction of the kinds of worker characteristics that might be relevant (see, for example, Agrell and Gustafson, 1996; Driskell et al., 1988). It may well be that whereas some people have strong needs to learn and respond positively to non-routine tasks, others may not. They may feel uncomfortable or insecure in a work environment that changes often, is ambiguous and full of variety. In this context, the personality characteristic ‘the tolerance for ambiguity’ (cf. Bowen et al., 1994) may be relevant for explaining different responses to non-routine work. People with a low tolerance for ambiguity may feel uncomfortable when confronted with non-routine situations, while those with a high tolerance do not see such situations as a threat, but rather as a challenge. Another construct that may be related to performance in non-routine environments is ‘self-efficacy’ (Bandura, 1993), which refers to the extent to which one believes that one is able to cope with a problem successfully. Some people have learned that they are able to solve (a certain area of) problems successfully and so, when confronted with failure, they will put a lot of effort into solving problems and into finding alternative strategies. Others, however, have learned that there is no relation between their efforts to solve (a certain area of) problems and the outcomes of these efforts. They assume that they have little or no control,

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

283

they will show little effort and will easily quit. In essence, these perceptions tend to have the character of a self-fulfilling prophecy: confidence in one’s own efficacy leads to more effort in a quantitative as well as in a qualitative sense, heightening the probability of success and thus fostering self-efficacy. A lack of confidence will inhibit effort and lower the chance of mastery and therefore (re)confirm the absence of control (Garber and Seligman, 1980). These two characteristics, tolerance for ambiguity and self-efficacy, are in line with what Senge (1990) states about the personal mastership of people in relation to learning. Individuals with a high level of personal mastership, see ‘reality’ as a friend, not as an enemy, so they are not afraid of ambiguity. They know what they are capable of and what they want, and have a great confidence in themselves. A third relevant trait is dispositional group loyalty, which refers to the need to be loyal to the team and to conform to group norms (cf. James and Cropanzano, 1994). Such loyalty contributes to what Senge has labeled collegiality. The three constructs just mentioned are interrelated, that is, persons who believe in their own problem-solving capacity will tolerate more ambiguity and be more committed to their own appraisals than to group norms (see, for example, Agrell and Gustafson, 1996). It is likely that self-efficacy and tolerance for ambiguity are positively related to performance in non-routine settings. With respect to group loyalty, however, we think that a curvilinear relation is more likely, especially in the case of team work: too much as well as too little loyalty will inhibit team performance. On the one hand, people who are not loyal to the group may not be willing to contribute to collective learning. On the other hand, those who feel a strong need to be loyal to their group have a strong tendency to conform to group norms and are less willing to demonstrate a deviant opinion. Therefore, they may contribute less to creativity and innovation. To link these characteristics to the STS principle of minimal critical specification, we think that STS should consider these characteristics in the case of selecting team members and composing teams. Besides, there are several ‘technical systems’ that support the selection and assessment of such personality traits (e.g. Jansen, 1997). With respect to these personality characteristics, we state the following proposition. Proposition 2. A high level of self-efficacy and tolerance for ambiguity and a moderate level of group loyalty among team members will promote team learning. 5.4. Practices: team characteristics With respect to team characteristics which are related to innovation, creativity and learning, we shall confine ourselves to three constructs: diversity, trust and power differentials within (temporal) teams. What we stated with respect to the characteristics of individual workers is also true here, that is, the variables we selected are only a selection out of all possibly relevant team characteristics (see, for example, Cummings, 1981; Sundstrom et al., 1990; Guzzo and Shea, 1992). 5.4.1. Diversity We will first consider diversity within teams with respect to abilities, cognitions and knowledge and second, we will consider diversity with regard to attitudes. If team members

284

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

are diverse with respect to abilities, cognitions and knowledge, the potential to come up with new ideas is larger and hence, there will be more opportunities to learn and a higher probability that new knowledge will emerge. On the other hand, communal knowledge and mental models, which we will find in ‘functional groups’, will facilitate communication and reduce coordination problems in teams (Volpe et al., 1996). Mishra and Spreitzer (1998) argue that a common frame of reference may improve the analysis and the appraisal of a situation, but will reduce the creative problem-solving behavior in a team. Jackson (1996) found that learning and creative processes are facilitated when teams are composed of persons with different (but somewhat overlapping) abilities. However, the more team members learn from each other, the more similar they will become with respect to their abilities and cognitions and the less innovative they will be. Thus, if innovation and creativity are the main issues, it is wise to limit the permanency of the composition of teams and to stimulate a certain level of turnover or rotation among teams, as participating in different groups may contribute to the creation and dissemination of knowledge. We have already related this to Senge’s fifth ‘discipline’ of systems thinking earlier in this section. With respect to attitudes, Jackson states that communality will improve communication and team performance. Social comparison theory (e.g. Suls and Miller, 1977) has shown that similarity with respect to attitudes reduces uncertainty (‘my view is normal’) and enhances mutual liking and trust, which facilitates the open exchange of knowledge and expertise. According to Senge, these are important conditions to break down defensive routines and enhance team learning. These points suggest that teams who have to deal with non-routine jobs should have common attitudes. However, common attitudes will stimulate the emergence of group norms which, as stated before, will regulate team members’ behaviors and force members to adhere to them (Tschacher and Brunner, 1995). This will inhibit deviant behavior and creativity (Neck and Manz, 1994) and may lead to what Janis (1972) has labeled ‘group think’. Group cohesion may promote success and success will foster group cohesion (e.g. Barry and Stewart, 1997), and this cycle in the end may result in a complacency that will undermine success. Particularly if creativity and innovation are important, the expression ‘never change a winning team’ seems to be untrue, which also implies that the optimal level of group cohesion is related to the routine/non-routine dimension of work (Adler and Docherty, 1998). These arguments also support the design practices we recommended. If we relate the above to the sociotechnical criterion, this STS principle would demand for a moderate level of variety with respect to cognitions, abilities and attitudes at a team level to support team learning. At an organizational level there should be more of such variety. Following the principle of minimal critical specification, directions for composing project teams can be deduced from the required level of variety. This brings us to the following propositions. Proposition 3a. A moderate level of variety at the team level with respect to knowledge, abilities, mental models and attitudes will promote team learning. Proposition 3b. A high level of variety at the organizational level with respect to knowledge, abilities, mental models and attitudes will promote organizational learning.

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

285

5.4.2. Trust Mutual trust among team members is an important condition that facilitates learning and successful performance when work is non-routine in nature (Jones and George, 1998). Trust, defined as an optimistic expectation on the part of an individual about the outcome of an event or the behavior of another person (Hosmer, 1995), seems to resemble the concept of collegiality in Senge’s work. If there is no trust, team members will be reluctant to contribute their expertise and knowledge, and in terms of Senge, they will demonstrate defensive routines, because they suspect others of being less cooperative. Members will only be willing to be open and contribute their knowledge and creative thoughts if they trust their fellow members. If one supposes that another team member is going to take credit for one’s effort, the motivation to be open and to contribute to the team task will disappear (Jones and George, 1998; Wicks et al., 1999). Moreover, in non-routine work environments, we mainly encounter professional workers, whose power is predominantly based on expertise. In the case of distrust, this may be an additional reason for them to stop contributing their expertise because they might expect to lose power. The concept of power will be further elucidated in the following sub-section. Relating the sociotechnical criterion to the concept of trust, we may conclude that there should be a high level of mutual trust in a team and thus not too much variety with respect to trust between workers. This brings us to the next proposition. Proposition 4. A high level of trust amongst team members will promote team learning. 5.4.3. Power Employees doing non-routine work will often, quite independently of management, make decisions with respect to the analysis of problems and to the working methods they apply. They will become a critical factor in the success of the firm and therefore, will have a lot of influence over the strategic choices of the organization and the choices made with respect to what and how to process. So there will be a lot of power on the operational level and thus, more leeway for an unequal distribution of power among members. An unequal distribution of power may form an obstacle for the best fulfillment of non-routine work and team learning (e.g. Jackson, 1996; Agrell and Gustafson, 1996). Senge refers to the issue of power inequality in terms of hierarchy and differences in rank. The more the powerful members dominate in problem-solving, the more the potential contribution of the others will be lost. Moreover, ‘expectations states theory’, which deals with ‘status organizing processes’, has shown that group processes tend to foster power inequalities, because the powerful will contribute more to the team output, which will reinforce their power (e.g. Berger et al., 1974). This will further undermine team learning. Without intervention, the powerful will likely become more powerful and those with little power will have even less. Technical systems, such as multi-media applications, may support such an intervention (see, for example, Jessup and Valacich, 1993; Muller, 1999). For example, we have experimented with team sessions in which members discussed issues via a computer network, which kept the knowledge, arguments and thoughts that members brought in anonymous. This enabled the existing skewed distribution of power within this team to be bypassed and it improved team learning. To relate the aforementioned, we use the following sociotechnical criterion.

286

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

Proposition 5. A low variety in power will enhance team learning. In this section, we considered organizations which strive for innovation and the creation of knowledge, besides efficiency, quality and flexibility. With respect to the practices deduced from the sociotechnical criterion, we think that STS should focus on design and process. Flexible structures will facilitate learning at the team as well as at the organizational level. A moderate level of diversity with respect to abilities, cognitions and attitudes at the team level will enhance team learning, whereas to foster organizational learning, a higher amount of diversity is desirable (see Fig. 1). In accordance with the principle of minimal critical specification, we have argued that in the case of non-routine work, it is important that there is much room for decision-making and collective learning activities at the operational level. This means that these processes themselves should be specified minimally. However, it is important to specify and create good learning conditions through managing diversity, mutual trust, and the distribution of power within teams, and through managing the selection of members and the composition of teams with respect to self-efficacy, tolerance for ambiguity and group loyalty. For the joint optimization of the technical and social system, we think that when learning and innovation are the leading competitive performance indicators, the social system (humans and their interpersonal relations) is of primary importance. However, the influence of technical systems, such as multi-media and group support systems, on learning and innovation is growing. Employees working in ‘learning organizations’ can mostly be typified as professionals doing challenging jobs which are highly intrinsically motivating. For example, they want to

Fig. 1. A learning approach to STS.

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

287

contribute to the solution of complex social problems, to the development of a knowledge field, to finding solutions for advanced technical or methodological problems, and they strive after personal success (Miner et al., 1994). Therefore, it seems reasonable to presume a high QWL. We will deal with this issue further in Section 6.

6. Discussion The main issue we have dealt with in this paper is the usefulness of STS as a diagnostic and design theory, one that can help managers select the appropriate practices given a particular set of performance indicators. We have done this by relating four patterns of performance indicators (efficiency, quality, flexibility, innovation) to three sociotechnical principles (the sociotecnical criterion, the principle of minimal critical specification and the principle of joint optimization). We have summarized this by means of a few keywords in Table 1. These keywords, such as ‘minimal variety’ or ‘no empowered jobs’, should not be interpreted as meaning that STS supports them, but that just applying STS principles in the case of specific performance indicators evokes such practices. In the balance of this section, we will deal with some implications of our paper and especially focus on how STS may mature further and contribute to recent and prospective developments. Before doing so, we first want to state that we confined our analyses to contingencies between demand patterns and management practices and ignored important issues such as the labor market, technological innovations, globalization, political factors and cultural differences. Therefore, the contingencies we described are non-deterministic. The same is true for STS as a design theory. In several sections of this paper we mentioned non-STS practices, some of which may nevertheless be easily incorporated in STS design applications, and which may well contribute to attaining specific performance indicators. However, we think that what makes STS different is that it emphasizes organizational performance as well as QWL in contrast to the other practices. If we consider performance indicators to be primarily an organization or management interest and QWL to be a worker interest, our arguments indicate that both these interests do not match very well if efficiency and quality are the only critical performance indicators. Moreover, in these circumstances STS practices seem to pertain mainly to the job level. When flexibility and/or innovation arises these interests seem to align much better. In the latter case, STS interventions focus on the job level as well as on the organizational level. Furthermore, in the case of innovation, STS may not only help to design structures but also help to design processes. These arguments may help to explain some of the conflicting research outcomes of STS. One may hypothesize that the reason why some studies show positive outcomes and other studies the opposite may be related to the outcomes being studied. Although we think that this is not the right place for an extensive review of former empirical research on STS, our previous research indicates such a link. In a car assembly shop where efficiency was the leading performance indicator, we found that there were hardly any possibilities of applying STS practices (Niepce and Molleman, 1996), while in a firm producing diodes, stacks and glass–metal where quality was a main issue besides efficiency, there was a need to delegate tasks such as machine set-ups, repair of simple machine breakdown, basic

288

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

289

equipment maintenance and quality control (Balkema and Molleman, 1999). We found similar results in a firm producing industrial glass (Hut and Molleman, 1998), while in a nursing context, we found a good basis for implementing STS practices (Molleman and Van Knippenberg, 1995). Advanced medical treatment and the growing assertiveness of patients, amongst other things, have changed the work of nurses. They have to deal increasingly with the unique needs of patients, making their work much more non-routine. Finally, in a study amongst IT professionals we found the right conditions to move towards a learning organization (Stam and Molleman, 1999). The historical perspective of Kumpe and Bolwijn (1994), which we depicted in Section 1, may explain why up to the 1970s, notwithstanding several more-or-less successful implementations, STS did not result in a fundamental and lasting breakthrough into organizational and job design. From a business or management point of view, there was no profitability in abolishing simple jobs and forcing back the division of labor. An increase in efficiency or quality could be expected more through non-STS related practices. Nevertheless, this historical viewpoint indicates a promising future for STS practices. At the end of this paper, we wish to build on the perspective that the move towards flexibility and innovation is dominant (the right part of Table 1). The first point we wish to touch on is the management of worker and team characteristics, which we claimed to be important in innovative firms. The constructs we discussed, such as trust and diversity, are rather vague and complex variables which are very difficult to manage. To create trust, for example, is a time-consuming and difficult process, while trust can very easily be undermined (Hosmer, 1995; Jones and George, 1998). Moreover, in the previous section we suggested curvilinear relations between some of these variables, such as diversity and group loyalty related to performance, so one may question where the optimum may lie. This may be further complicated when the optimum of the one variable (e.g. diversity) differs from that of another one (e.g. group cohesion), or when different variables influence performance in opposite ways. The need for diversity, for example, leads to teams whose members have different domains of knowledge, while in order to facilitate the emergence of trust, one would prefer members to share the same domain (Mayer et al., 1995). A certain level of diversity may stimulate the development of new insights and frames of reference, while at the same time it may impair trust. Another complicating factor is the dynamic side of these constructs. When people work together, trust and group cohesion may grow, but diversity may diminish and group norms may become too coercive. When is the right moment to change a team? An additional facet of dynamism is that when teams are designed and structured along such rather vague constructs as trust and diversity, there will be much leeway for each member to have his or her own interpretation of these structures. Moreover, in teams in which creative and learning processes are important, workers will influence each other’s perceptions and will be continuously involved in a cycle of experimentation and learning. Driven by creativity and improvisation, they may regularly give new interpretations to structures, work design and procedures. This means that the design and structure of teams evolves and changes continuously over time (Weick, 1993; Moorman and Miner, 1998). In sum, we think that all these practical questions and complications should direct further research.

290

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

A second issue is related to the classical STS aim of striving for higher levels of QWL. In the previous section, we concluded that employees working in innovative and knowledge creating firms will mostly have appealing jobs which are intrinsically motivating, resulting in a good QWL. The other side of the coin is that intrinsically motivated people may create their own work and continuously search for new challenges, which may make their jobs endless and the workload boundless. The feeling that the job is never done may cause stress or even burnout and a low perceived QWL. This raises another important research question on how to manage intrinsic drives and at the same time to protect workers from such pitfalls. A last question on which we can only speculate is what will come after innovation and learning and what should be the response of STS to future developments in performance indicators? We wish to touch briefly on one development. We think that the role of the customer in the processing of goods and services will change dramatically and that customer care will become a new critical performance indicator. Customer care refers to the establishment of enduring relationships in which the dialogue with the customer becomes the main issue (Edvardsson et al., 1994). It means that the customer (e.g. a university student) participates in the team as a resource, co-producer, buyer, user and as a product of the team activities itself, in the sense of being the key outcome of transformation activities (Lengnick-Hall, 1996). Questions may emerge such as ‘how to manage clients in their different and various roles?’, ‘what is the position of the client in a team?’, ‘how to develop trustful relationships when the client is both co-producer and buyer?’. We think that the research agenda in the field of STS should include projects which focus on the involvement and participation of clients in flexible teams. We will end with a final comment. An STS principle we did not deal with in this paper is ‘the principle of incompletion’, which states that the design process is an ongoing process and will never be finished (Cherns, 1987). We think that this principle is also valid for the development of STS theory itself.

Acknowledgements We wish to thank three anonymous reviewers and the editors of this special issue for their helpful comments on two earlier versions of this paper.

References Adler, P.S., Cole, R.E., 1993. Designed for learning: a tale of two auto plants. Sloan Management Review 85–94. Adler, N., Docherty, P., 1998. Bringing business into sociotechnical theory and practice. Human Relations 51, 319–345. Agrell, A., Gustafson, R., 1996. Innovation and creativity in work groups. In: West, M.A. (Ed.), Handbook of Work Group Psychology. Wiley, New York. Argyris, C., Schön, D.A., 1996. Organizational Learning. II. Theory, Method, and Practice. Addison-Wesley, Reading, MA. Ashby, W.R., 1969. Self-regulation and requisite variety. In: Emery, F.E. (Ed.), Systems Thinking. Penguin Books, Harmondsworth.

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

291

Balkema, A., Molleman, E., 1999. Barriers in the development of self-organizing teams. The Journal of Managerial Psychology 14, 134–149. Bandura, A., 1993. Perceived self-efficacy in cognitive development and functioning. Educational Psychologist 28, 117–148. Barry, B., Stewart, G.L., 1997. Composition, process and performance in self-managed groups: the role of personality. Journal of Applied Psychology 82, 62–78. Benders, J., 1993. Optional Options: Work Design and Manufacturing Automation. Avebury Ashgate Publishing, Aldershot. Berger, J., Conner, T.L., Fisek, M.H. (Eds.), 1974. Expectation States Theory: A Theoretical Research Program. Winthrop, Cambridge, MA. Bowen, J., Qiu, Z., Li, Y., 1994. Robust tolerance for ambiguity. Organizational Behavior and Human Decision Processes 57, 155–165. Burbidge, J.L., 1992. Production flow analyses for planning group technology. Journal of Operations Management 10, 5–27. Cherns, A., 1987. Principles of sociotechnical design revisited. Human Relations 40, 153–162. Conti, R.F., Warner, M., 1993. Taylorism new technology and just-in-time systems in Japanese manufacturing. New Technology Work and Employment 8, 31–42. Cook, S., Yanow, D., 1993. Culture and organizational learning. Journal of Management Inquiry 2, 373–390. Crosby, P.B., 1979. Quality is free. The art of making quality certain. McGraw-Hill, New York. Cummings, T.C., 1981. Designing effective work groups. In: Nystrom, P.C., Starbuck, W.H. (Eds.), Handbook of Organizational Design, Vol. 2. Oxford University Press, New York. Cummings, T., Blumberg, M., 1987. Advanced manufacturing technology and work design. In: Wall, T.D., Clegg, C.W., Kemp, N.J. (Eds.), The Human Side of Advanced Manufacturing Technology. Wiley, London. Daft, R.L., Huber, G., 1987. How organizations learn: a communication framework. In: Bacharach, S.B. (Ed.), Research in the Sociology of Organizations 5, 1–36. Dankbaar, B., 1997. Lean production: denial confirmation or extension of sociotechnical systems design. Human Relations 50, 567–583. Delbridge, R., Turnbull, P., Wilkinson, B., 1992. Pushing back the frontiers: management control and work intensification under JIT/TQM factory regimes. New Technology, Work and Employment 7, 97–106. De Sitter, L.U., Vermeulen, A.A.M., Van Amelsvoort, P., Van Geffen, L., Van Troost, P., Verschuur, F.O., 1986. Het flexibele bedrijf: Integrale aanpak van flexibiliteit, beheersbaarheid, kwaliteit van de arbeid en produktie-automatisering (the flexible firm: an integral approach to flexibility, efficiency, quality of working life and manufacturing technology). Kluwer Academic Publishers, Deventer. De Sitter, L.U., Den Hertog, J.F., Dankbaar, B., 1997. From complex organizations with simple jobs to simple organizations with complex jobs. Human Relations 50, 497–534. Driskell, J.E., Hogan, R., Sales, E., 1988. Personality and group performance. Review of Personality and Social Psychology 14, 91–112. Dunphy, D., Bryant, B., 1996. Teams: panaceas or prescriptions for improved performance? Human Relations 49, 677–699. Ebeling, A.C., Lee, C.Y., 1994. Cross-training effectiveness and profitability. International Journal of Production Research 32, 2843–2859. Edvardsson, B., Thomasson, B., Ovretveit, J., 1994. Quality of Service: Making it Really Work. McGraw-Hill, Berkshire. Emery, F.E. (Ed.), 1969. Systems Thinking. Penguin, London. Emery, E.E. (Ed.), 1978. The Emergence of a New Paradigm of Work. Centre for Continuing Education, Australian National University, Canberra. Fazakerley, G.M., 1976. A research report on the human aspects of group technology and cellular manufacture. International Journal of Production Research 14, 123–134. Francis, A., Winstanley, D., 1988. Organizing professional work: the case of designers in the engineering industry in Britain. In: Pettigrew, A.M. (Ed.), Competitiveness and the Management Process. Blackwell (Basil), Oxford. Galbraith, J., 1973. Designing Complex Organizations. Addison-Wesley, Reading, MA. Garber, J., Seligman, M.E.P. (Eds.), 1980. Human Helplessness, Theory and Applications. Academic Press, New York. Golembiewski, R.T., 1995. Managing Diversity in Organizations. The University of Alabama Press, Tuscaloosa.

292

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

Guzzo, R.A., Shea, G.P., 1992. Group performance and intergroup relations in organizations. In: Dunette, M.D., Hough, L.M. (Eds.), Handbook of Industrial and Organizational Psychology, Vol. 3. Consulting Psychologists Press, Palo Alto, CA. Hackman, J.R., Oldham, G.R., 1980. Work Redesign. Addison-Wesley, Reading, MA. Hauser, J.R., Clausing, D., 1988. The house of quality. Harvard Business Review 67, 63–73. Herbst, P.G., 1974. Socio-Technical Design: Strategies in Multidisciplinary Research. Tavistock Publications, London. Hitomi, K., 1993. Manufacturing technology in Japan. Journal of Manufacturing Systems 12, 209–215. Hosmer, L.T., 1995. Trust: the connecting link between organizational theory and philosophical ethics. Academy of Management Review 20, 379–403. Huber, V.L., Brown, K.A., 1991. Human resource issues in cellular manufacturing: a sociotechnical analysis. Journal of Operations Management 19, 138–159. Hut, J.A., Molleman, E., 1998. Empowerment and team development. Team Performance Management Journal 4, 53–66. Jackson, S.E., 1996. The consequences of diversity in multidisciplinary work teams. In: West, M.A. (Ed.), Handbook of Work Group Psychology. Wiley, New York. James, K., Cropanzano, R., 1994. Dispositional group loyalty and individual action for the benefit of an ingroup: experimental and correlational evidence. Organizational Behavior and Human Decision Processes 60, 179–205. Janis, I.L., 1972. Victims of Groupthink. Houghten Mifflin, Boston. Jansen, P.G.W., 1997. Assessment in a technological world. In: Anderson, N., Herriot, P. (Eds.), International Handbook of Selection and Assessment. Wiley, New York. Jessup, L.M., Valacich, J.S. (Eds.), 1993. Group Support Systems — New Perspectives. Macmillan, New York. Jones, G.R., George, J.M., 1998. The experience and evolution of trust: implications for cooperation and teamwork. Academy of Management Review 23, 531–546. Legge, K., 1991. Human resource management: a critical analysis. In: Storey, J. (Ed.), New Perspectives in Human Resource Management. Routledge, London. Lengnick-Hall, C.A., 1996. Customer contributions to quality: a different view of the customer-oriented firm. Academy of Management Review 21, 791–824. Kolb, D.A., 1984. Experiential Learning. Prentice-Hall, Englewood Cliffs. Krafcik, J.F., 1988. Triumph of the lean production system. Sloan Management Review 30, 41–52. Kuipers, H., Van Amelsfoort, P., 1990. Slagvaardig organiseren: inleiding in de sociotechniek als integrale ontwerpleer (Organizing in an Effective Way: Introduction to the Sociotechnical Systems Theory as an Integral Design Theory). Kluwer Academic Publishers, Deventer. Kumpe, T., Bolwijn, P.T., 1994. Toward the innovative firm — challenge for R&D management. Research Technology Management 37, 38–45. Manz, C.C., 1992. Self-leading work teams: moving beyond self-management myths. Human Relations 45, 1119– 1140. Manz, C.C., Stewart, G.L., 1997. Attaining flexible stability by integrating total quality management and socio-technical systems theory. Organization Science 8, 59–70. Mayer, R.C., Davis, J.H., Schoorman, F.D., 1995. An integral model of organizational trust. Academy of Management Review 20, 709–734. Miner, J.B., Crane, D.P., Vandenberg, R.J., 1994. Congruence and fit in professional motivation theory. Organization Science 5, 86–97. Mintzberg, H., Quinn, J.B., 1992. The Strategy Process: Concepts and Contexts. Prentice-Hall, London. Mishra, A.K., Spreitzer, G.M., 1998. Explaining how survivors respond to downsizing: the roles of trust, empowerment, justice and work redesign. Academy of Management Review 23, 567–588. Molleman, E., Van Knippenberg, A., 1995. Work redesign and the balance of control within a nursing context. Human Relations 48, 795–814. Moorman, C., Miner, A.S., 1998. Organizational improvisation and organizational memory. Academy of Management Review 23, 698–723. Morgan, G., 1986. Images of Organizations. Sage, Beverley Hills. Muller, P., 1999. Teambased conceptualization of new products: creating shared realities using information technological support. Dissertation. University Press, Groningen.

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

293

Neck, C.P., Manz, C.C., 1994. From groupthink to teamthink: toward the creation of constructive thought patterns in self-managing work teams. Human Relations 47, 929–952. Niepce, W., Molleman, E., 1996. Characteristics of work organization in lean production and sociotechnical systems: a case study. The International Journal of Operations and Production Management 16, 78–91. Niepce, W., Molleman, E., 1998. Work design issues in lean production from a sociotechnical systems perspective neo-Taylorism or the next step in sociotechnical design. Human Relations 51, 259–287. Parker, M., Slaughter, J., 1992. Choosing Sides, Unions and the Team Concept. South End Press, Boston. Pasmore, W.A., 1988. Designing Effective Organizations: The Sociotechnical Systems Perspective. Wiley, New York. Pasmore, W.A., 1995. Social science transformed: the social–technical perspective. Human Relations 48, 1–21. Pava, C., 1986. Redesigning sociotechnical systems design: concepts and methods for the 1990s. The Journal of Applied Behavioral Science 22, 201–221. Schneider, B., 1994. A service perspective: towards a custom-focused HRM. International Journal of Service Industry Management 5, 64–76. Schonberger, R., 1982. Japanese Manufacturing Techniques: Nine Hidden Lessons in Simplicity. Free Press, New York. Senge, P.M., 1990. The Fifth Discipline; The Art and Practice of the Learning Organization. Doubleday, New York. Shingo, S., 1985. A Revolution in Manufacturing: The SMED System. Productivity Press, Cambridge. Shingo, S., 1986. Zero Quality Control: Source Inspection and the Poka–Yoke System. Productivity Press, Cambridge. Stam, M., Molleman, E., 1999. Balancing the demand for and supply of IT-specialists: towards a learning organization. The International Journal of Manpower 20, 375–387. Suls, J.M., Miller, R.L. (Eds.), 1977. Social Comparison Processes: Theoretical and Empirical Perspectives. Hemisphere, Washington, DC. Sundstrom, E., De Meuse, K.P., Futrell, D., 1990. Work teams: applications and effectiveness. American Psychologist 45, 120–133. Suresh, N.C., 1992. Partitioning work centers for group technology: analytical extension and shop-level simulation investigation. Decision Sciences 23, 267–290. Suresh, N.C., Kay, J.M. (Eds.), 1998. Group Technology and Cellular Manufacturing; State-of-the-Art Synthesis of Research and Practice. Kluwer Academic Publishers, Boston. Suzaki, K., 1993. The New Shop Floor Management: Empowering People for Continuous Improvement. Free Press, New York. Thompson, J.D., 1967. Organizations in Action. McGraw-Hill, New York. Trist, E., Bamforth, K.W., 1951. Some social and psychological consequences of the Longwall method of coal-getting. Human Relations 4, 6–24. Trist, E.L., Murray, H. (Eds.), 1993. The Social Engagement of Social Sciences: A Tavistock Anthology: The Sociotechnical Perspective, Vol. II. University of Pennsylvania Press, Philadelphia. Tschacher, W., Brunner, E.J., 1995. Empirische studien zur dynamik von gruppen aus der sicht der selbstorganisationstheorie. Zeitschrift fur Sozialpsychologie 16, 78–91 (in German). Turnbull, P.J., 1988. The limits to Japanisation just-in-time, labor relations and the UK automotive industry. New Technology Work and Employment 3, 7–19. Van der Zwaan, A.H., 1975. The socio-technical systems approach: a critical evaluation. International Journal of Production Research 13, 149–163. Van Eijnatten, F.M., 1993. The Paradigm that Changed the Work Place. Van Gorcum, Assen. Van Eijnatten, F.M., Van der Zwaan, A.H., 1998. The Dutch IOR approach to organisational design, an alternative to BPR. Human Relations 50, 290–318. Vogel, E.F., 1979. Japan as Number One: Lessons for America. Harvard University Press, Cambridge. Volberda, H.W., 1996. Toward the flexible firm: how to remain vital in hypercompetitive environments. Organization Science 7, 359–374. Volpe, C.E., Cannon-Bowers, J.A., Sales, E., Spector, P.E., 1996. The impact of cross-training on team functioning: an empirical investigation. Human Factors 38, 87–100. Wall, T.D., Clegg, C.W., Kemp, N.J. (Eds.), 1987. The Human Side of Advanced Manufacturing Technology. Wiley, London.

294

E. Molleman, M. Broekhuis / J. Eng. Technol. Manage. 18 (2001) 271–294

Weick, K.A., 1993. Organizational redesign as improvisation. In: Huber, G.P., Glick, W.H. (Eds.), Organizational Change and Redesign: Ideas and Insights for Improving Performance. University Press, Oxford. Wicks, A.C., Berman, S., Joners, T.M., 1999. The structure of optimal trust: moral and strategic implications. The Academy of Management Review 24, 99–116. Williams, K., Haslam, C., Williams, J., Cutler, T., Adcroft, A.S., Johal, S., 1992. Against lean production, economy. Economy and Society 2, 321–354. Womack, J.P., Jones, D.T., Roos, D., 1990. The Machine that Changed the World. Rawson Associates, New York.