The socio-technical approach to work organisation. An essential element in Quality Management Systems

This study analyses the techniques and models of quality management from the perspective provided by the design of socio-technical systems, proposed as an alternative paradigm to the Taylorist model of work organisation. This analysis seeks to show how the absence of a widely accepted alternative paradigm to that of Taylorism has favoured the continued use of Taylorism through the techniques and methodologies of quality that have been developed and applied in industry in recent years.

Keywords: socio-technical systems; quality and organisational change

The enormous success experienced by the US in its industrialisation at the beginning of the 20th century, using the Taylorist model, caused this model to spread rapidly throughout the developed world, and to become the true management paradigm for companies and organisations during the century.

It is true that the mechanistic Taylorist model did not fit people's real behaviour, and one proof of this poor fit are the results – widely published in textbooks and journals – obtained by Elton Mayo experiments at the Hawthorne plant.

Probably the most widely disseminated idea from these studies by Mayo was the discovery that people modify their behaviour when observed. It is also likely that this had already been discovered as a consequence of the measurement of the duration of activities made in the many work studies that were already common in industry at that time. Undoubtedly there were already cases of workers simulating activity when they were being observed, but the publication of these studies, linked as they were to the (nearly) scientific experiments of Mayo, contributed to the success of these ideas.

Nevertheless, the success of Mayo's experiments was based not so much on the discovery that people's behaviour varies when they are observed, but on the direct consequence of this discovery: that motivation can modify workers' behaviour and so can affect the productivity achieved by workers.

This discovery opened the way to compensating workers for the most evident drawbacks of the application of the Tayloristic model. This early interest shown by industry in looking for incentives that would get people to work in a mechanistic way is tangible proof that the application of the Taylorist paradigm had already begun to show its inadequacies during the very period in which it was at its height.

The truth is that the aspect of motivation and the possibility of using it to offset the drawbacks in the Taylorist system was of much greater interest to industry than other findings from the Hawthorne experiments, and helped to establish Mayo's scientific prestige. He went on to become the founder of a school of management thought – the Human Relations School – which still exists.

Nevertheless, from the description of the Hawthorne experiments, which the collaborators of Mayo, Roethlisberger and Dickson [17] carried out, one can deduce that the motivation reached in the experiments was external to the work itself and does not correct the problems caused by the application of the Taylorist approach. Their effects are short lasting and, once the initial surprise passed and the workers realised that these changes were not intended to improve their working conditions, the motivation disappeared.

The Taylorist model of an organisation's lack of fit has grown in stature (Prida, [15]) throughout the 20th century due to the increasing contradiction between:

• • the maintenance of a mechanistic model that treats people as simple cogs in the technical process, and condemns them to carrying out boring and meaningless tasks as the 'one best way' of achieving maximum efficiency; and
• • the raising of people's expectations as a result of increasingly better training and education, at least in the developed world.

And this poor fit becomes more obvious with each passing day in both manufacturing and the service industry, as shown by statistics on absenteeism (active or passive) and on defects in the quality of products or services (Sitter et al., [19]).

The prestige of the Taylorist paradigm fell increasingly at the end of the 20th century and the newest motivation theories have not changed this fact (Sprenger, [20]). Today, only a few managers would dare to confess that they still apply this paradigm, although the fact is that many of the basic features of the Taylorist approach are still applied today, more than a century after its appearance.

As we see later, attempts to establish a new paradigm that would be different from the Taylorist model, via the socio-technical design of production systems, have been eclipsed by the enormous diffusion of techniques and methodologies in which all references to either paradigm have been omitted.

It is probably the absence of a widely accepted alternative paradigm to that of Taylorism that is still favouring the increasing appearance of new theories and methods of management and organisation, where it is common to see features that are clearly Taylorist alongside others that are quite different.

One example of these techniques is quality management systems, where one can see that the lack of explicit references to Taylorism and to the design of socio-technical systems leads to applications of these techniques where one often finds a mixture of both paradigms.

This paper does not pretend to carry out an exhaustive review of models of work organisation applied in industry from the onset of Taylorism until today. Our purpose is only to make evident the continuity of the Taylorist paradigm in many of the newer techniques of quality management that have been developed and applied in industry since Taylorism first appeared, even though there is increasing evidence of a lack of fit of this paradigm.

As we said before, the problems caused by an all-out application of the Taylorist model began to be quickly noticed. However, it would not be until the 1950s that some researchers of the British Tavistock Institute would make this evident in a series of studies in several English coal mines.

Here, after an intense process of mechanisation in which, following the Taylorist model, a strict definition of tasks was established, productivity – contrary to all expectations – dropped. The researchers presented an analysis of the probable causes of this situation, in which they questioned the Taylorist paradigm being the 'one best way' to achieve high productivity.

The Institute's researchers pointed out that the process of mechanisation had destroyed the traditional structure of small groups of miners who carried out all the tasks related to coal mining, replacing it with specialists who worked independently in highly specialised jobs.

The solution suggested by the researchers was a new form of work organisation, which they called the Design of socio-technical systems, inspired by systems theory, biology, logic, and cybernetics, from which they adopted concepts and theories that were later modified on the basis of practical experience.

• • A holistic system, which makes it possible to look at the overall situation, adopting an integrated view of a production system made up of a technical system and a social subsystem in continuous interaction.
• • An open system, where it is necessary to pay attention to the relations between its elements and between the system and the environment.
• • The self-regulation, which is the basis of work groups.

Conceptually, the new model means a change in the design of work organisation. Traditionally, in the Taylorist model, engineers decide what form of organisation is best, based on the demands of the technical system, and without taking into account the relationship between the technology and people. The theory of socio-technical systems, as opposed to the Taylorist model, proposes the necessity of establishing a joint design of the technical system and the social system.

To do this, the socio-technical model proposes the analysis of both the social and the technical systems, on an equal basis, and the study of the relationship between them. This is because consideration that the proper performance of the system depends more on how its parts interact than on how they work independently and the optimum performance of a system cannot be reduced to the sum of the optimum performance of its parts (Trist, [23]).

This process will allow each company to develop a new model of work organisation for the joint optimisation of the social subsystem and the technical subsystem, which takes into account the significant influence of the present and future demands of both subsystems in relation to the environment.

Taking the social subsystem and the influence of the environment into account in the design of a work organisation this represents a major change in the paradigm versus the traditional Taylorist approach. This change in the role of the workers in the company constitutes a basis of the new paradigm, represented by the following design principles (Trist, [23]):

• • The work system – which is made up of a group of activities that make it work as a whole – is now the basic building block, instead of the individual tasks into which it was previously broken down.
• • In accordance with this, the work group becomes the key element, rather than the individual worker.
• • Thus, it is possible for the internal regulation of the system to be done by the group, instead of the workers being overseen externally by supervisors.
• • The basic organisational philosophy is based on a redundancy of functions instead of a redundancy of parts, as this tends to develop multiple skills in each individual and to increase the group's repertoire of responses and its flexibility.
• • This principle favours discretionary behaviour over set work roles.
• • In addition, it treats the individual as a complement to the machine, rather than as an extension of it.
• • It seeks an increase in variety for both the individual and the organisation, versus its reduction in the bureaucratic model.

Several years later, Emery Trist [23] added to these principles new criteria for the design of individual work roles, relating to their content, which should allow each person an optimum level of variety, opportunities for learning, etc.

The implantation of these new principles in organisations is carried out mainly through work groups. While they are not meant to be a standard solution within the model of socio-technical design, the possibilities that they offer for cooperation and coordination among people have made them a key element in work organisation for defining more democratic and participative work structures.

The recognition of the importance of the social system in work organisation, in the socio-technical approach, in contrast to other approaches, such as Taylorism, leads to a replacement of individual work methods with group work methods. The multi-functionality of the work groups allows for an interchange of jobs, reduces the need for coordination between workers and bureaucracy, and finally increases workers' autonomy, especially in making decisions directly related to the activities that they carry out.

Thorsrud [22] establishes, as one of the most important findings of these experiments with groups, the concept of self-directed learning, because when one increases the variation in work content, participation and commitment in decision-making also increase, and interaction between people is encouraged.

Nevertheless, to ensure the effectiveness of the participative process, this change in the structure of decision-making from the traditional system to the socio-technical system should not be interpreted as just a more-or-less participative restructuring of tasks at the shop floor level, but should bring with it more radical modifications in the organisation, such as changes in supervisory roles (Prida, [14]) or in payment systems (Pasmore, [13]).

In addition, it must be remembered that, on occasion, conflicts can arise between groups, since each group can make changes in their work process that create difficulties for other workers. Thus, mechanisms of coordination between groups will always be necessary in order to centre their work on the outcome of the process and on the product. Even so, the design principles of the socio-technical approach seek to reduce or minimise these relations between groups, by assigning each of them an independent unit of operation where the transformation of the product is done.

The definition of these operational units, where product transformation takes place, once again implies major differences in the paradigm of the socio-technical model with respect to Taylorism:

• • The new focus is on the product and not the task, on the outcome of the action, on the client, and not on the action itself.
• • The objective is control of the product by the workers, instead of control of the workers.

This view allows us to see once again the possibilities afforded by worker involvement and participation, from a different perspective. First of all, the workers share a common work language – the product – which makes them strive to achieve this one goal instead of simply doing the tasks that are assigned them without knowing what these are for. Secondly, they are in the best position to control the process while they are working: they know how the process works, what its key variances are, and how to keep these under control.

The involvement and participation of the workers not only improves their quality of life, but has also proved to be very useful for facilitating organisational flexibility and improving productivity and product quality. All this leads to both worker and management satisfaction (Molleman & Broekhuis, [12]).

Even so, despite the good results obtained in many experiments carried out in different companies in various developed countries,[1] based on methodological advances achieved by a group of researchers whose work was orientated towards action research (Ziegenfuss, [24]), the truth is that new paradigms that would be an alternative to the Taylorist approach have not been widely disseminated.

It could be said that the socio-technical paradigm is not yet fully developed, especially in technical and methodological terms, and because of this its success when put into practice has been strongly dependent on the ability of directors and researchers to understand the socio-technical system where it is being applied.

However, this is just the type of difficulty that Kuhn [11] said is faced by a new, emerging, and still incompletely developed paradigm. In the early stages, it is making its way in opposition to an old paradigm, which – in spite of its defects and undeniable contradictions that appear with increasing frequency – is still deeply rooted in both business as well as technological and scientific culture. All of this is more than enough to explain this slow development of the socio-technical paradigm,

However, the turbulent environment in which the industry and the developed societies are immersed has, since the 1980s, eclipsed the advances of new paradigms with a multitude of new methods, theories and management techniques of general applicability that promise quick results.

In the 1980s, the development of the socio-technical work and organisation paradigm underwent an enormous upheaval with the massive arrival of Japanese products, which displaced many Western products in their normal markets, and stunned the majority of Western companies.

In this period, both scientists and managers tried to find an explanation for the phenomenon of Japanese industrial success. Some began to focus on the 'most visible' part of its production system – its technology – but soon realised that Western industry was at this point still technologically far superior to Japanese industry. Others thought that Japanese workers accepted working conditions that could not feasibly be applied in the West. Finally, another group – probably the majority – thought they could see the reason behind Japanese success as being their superior methods, linked fundamentally to the Just in Time (JIT) approach (Schonberger, [18]), and their quality methods (Deming, [4]). Accordingly, many Western companies began to apply, on a massive scale and with varying success, many of the techniques and methods that were being applied in Japan.

In order not to sell defective products, Japan's industries, in the 1950s, realised that it was necessary to produce perfect goods from the start of the production process. Consequently, even though quality control was orientated towards product inspection, it soon evolved into prevention as a way of controlling those production factors, which produced defective products. It also involved all the functions of the company, not just manufacturing, in the quality process. Each worker is responsible for product quality, process improvement, and waste reduction. There are no departments dedicated to quality control or process engineering.

The way in which Japanese companies put these principles into practice was just as important as the principles themselves. First of all, the whole organisation was involved and committed. Second, as they had greater responsibility, workers needed more education and training, which led Japanese companies to dedicate considerable resources to training. Third, Japanese companies used classic principles of experimentation and measurement to judge whether alternative methods were better or not (Imai, [8]).

These concepts have helped Japanese companies to generate a type of thinking that is centred on the process, and to develop strategies designed to assure continuous improvement, which would involve people from all levels of the organisation: the KAIZEN (Figure 1).

Graph: Figure 1. The KAIZEN umbrella (Imai, [8]).

Probably this surprise at the success of Japanese companies in creating quality products, and the tremendous impact that this had on competitiveness, is one of the reasons why many of the Japanese quality methods have been those most widely disseminated in Western companies. Continuous improvement and quality management systems offer today techniques and models that are a common reference for most companies in the developed world, regardless of their size or the sector in which they do business.

However, all of these techniques and tools, which were clearly part of an integrated approach, and whose development and introduction are quite similar in many ways to the socio-technical focus that we have seen in the previous section, arrived in the West as independent and unrelated fragments.

This fragmentation has facilitated the quick dissemination of many of these techniques, but has led to them being seen by many companies as mere ways of improving productivity which can be applied pragmatically, and because of this many of them have become authentic management fads (Jackson, [10]).

This pragmatism, in applying techniques without an analysis of the impact of these techniques in the socio-technical systems work where they are applied, is the cause of many of the failures that have resulted from the introduction of these methods and techniques in many enterprises.

By way of example, below are some of the underlying contradictions present in the most popular models of quality management:

• • Continuous improvement is today one of the elements that all models referring to quality have in common, along with the well-known Deming Cycle or PDCA Cycle (Plan, Do, Check, Act). Continuous improvement is today a method that is referred to in all the quality management textbooks and which can be found in all the reference manuals. This method has proved to be an important tool due to both its simplicity and its efficacy.  Nevertheless, many companies have used this tool as a way of tapping into employees' creative reserves, using a type of 'delegation of the directive role' for the group learning, which must take place in the organisation. They thus encourage truly brilliant ideas, but forget the need for creative employees who consider themselves responsible for constantly helping to resolve problems, and thus there is a need for the integration of creativity and continuous improvement in the day-to-day management process.  Only in this way can the PDCA cycle be transformed into an experimental method similar to those used in the socio-technical approach (Taylor & Felten, [21]). Although this method is quite distant from the experimental methods of the physical sciences (Checkland, [3]), its use requires an even greater discipline and patience, since the people and their attitudes toward work are what make the data credible.  Nevertheless, it is not infrequent to see cases in which the PDCA cycle is applied to justify previously designed solutions, or cases in which management takes a passive attitude toward the experimentation, expecting ideas to emerge from the workers by spontaneous generation, rather than dedicating resources and time to planning the experimentation, etc.  The consequences of this weak reference to the work paradigm are undoubtedly clear to anyone who looks beyond mere data:
• • proposals almost always refer to others' areas of competence;
• • proposals proliferate, but they are very seldom really acted upon;
• • evaluation replaces the true planning, which would try to bring about action;
• • management's moralising calls for proposals to be made show those proposals are not standard practice in the company;
• • proposals are the exception, not the rule, and the goal that the employee seeks in making proposals is the 'prize' offered;
• • and so on.
• • The process approach, along with continuous improvement, dominates current quality methods, whether in the quality systems established according to ISO 9000:2000 norms, or EFQM, and is on its way to being a new management fad. Certainly, the identification of the existence of processes in a company is none other than the acceptance that a company is a system made up of different sub-systems, which interrelate with the aim of achieving common goals. Nevertheless, despite the acceptance of the company as a system, recognising it as a socio-technical system is much less common and often contradictory.  The ISO (AENOR, [1], [2]) model simply ignores it, although at least it allows management to choose the most appropriate organisational paradigm. Without a doubt, the preparation and observance of clear and detailed procedures is the basic element of this system; however, two opposing options can fit within it:
• a. Procedures designed to limit the actions of individuals or groups with the aim of assuring that the procedures established by management are strictly followed – a true rediscovery of the Taylorist model, more than a century after the birth of this paradigm (Hammer & Champy, [7]).
• b. Procedures designed to initiate a rigorous process of experimentation, which will allow new ideas to be put into practice and their effectiveness to be checked: a type of action research methodology.
• It is true that the quality 'philosophy' underlying much of the continuous improvement methodology applied in early Japanese experiences is closer to the second option than to the first. Nevertheless, either of the above options is compatible with an ISO certification, as long as data can be obtained that demonstrate some type of improvement in the system.
• Other groups of reference models in quality management such as EFQM ([5], [6]) Malcolm Baldrige, etc., formally recognise that a company is a system, and the requirement to keep in mind the relation of the company with different stakeholders.
• They establish a point model to evaluate each part of the system, without taking into consideration the interactions among the various stakeholders which are operating in an organisation (it is well known that, in a system, the sum of the parts cannot explain the behaviour of the whole). And they are centred on evaluation, neglecting the other phases established in Deming's cycle, which are necessary for rigorous experimentation.
• These models creates a certain amount of confusion by proposing a focus on processes that is quite a revolutionary for most of today's organisations with their traditional organisational structures, and instead of proposing a management model for organisational change (participative, authoritarian, consensual, etc).
• A clue for this short-term focus is the constant reference to the leadership as the medium to reach organisational change. The leadership does not propose any way to change the organisation, it simply acts as a new kind of motivation element necessary to get people's acceptance of the proposed model.
• In this situation, it is not rare to find applications of this method as a new type of mechanistic approach, a kind of 'new one best way' applicable to organisations. Thus, we should not be surprised when we find applications that only look to demonstrate a quick success (marketable to different stakeholders) and employ this model as the universal 'justification' of excellence applicable to all cases.
• In short, these models and all their ambiguity can be used as easily to protect badly-run organisations that wish to obtain a seal of approval, as to reflect seriously on how the existing socio-technical system works, without this necessarily resulting in a better or worse quality model.

It is not easy today to get true information of the results of the applications of these techniques in a literature more interested in reflecting the successes than the failures. An exception is the well-known case of the Wallace Company, which declared bankruptcy a few years after it received the Malcolm Baldrige Quality Award (Ivey, [9]); however, in our experience in academic (Rey García, [16]) and consulting applications, we think that it is only the tip of a big iceberg.

More than a century after its invention, Taylorism is still in good health in many of today's organisations, but as time passes the poor fit of the mechanistic model to the people of developed countries has kept growing. The analysis made in this article demonstrates that, despite its current formal discrediting, Taylorism is still present in the application of many of the new organisational solutions that are meant to be novel. In this situation, the subterfuges used in order to continue applying the Taylorist approach, form an inherent part of the paradigm.

Elton Mayo was perhaps the inventor of the first subterfuge – the motivation approach (Sprenger, [20]). But although he received little publicity for it, Elton Mayo himself discovered in his experiments at Hawthorne that these subterfuges do not solve the problems created by the Taylorist approach, but only hide their effects.

However, they are not able to hide them for very long and, in addition, this occultation normally produces new problems and even greater conflicts in the socio-technical systems where it is employed. These little-publicised discoveries of Mayo are still valid today. More than 50 years have passed since the experiments at Hawthorne, and the type of methods used to hide the problems caused by the Taylorist approach have changed (although sometimes the same old methods are still applied). Nevertheless, Mayo's little-publicised discoveries are still valid today.

Finally, one can still draw important conclusions by reflecting on Mayo's experiments, and his discovery that motivation affects workers' behaviour. This conclusion could be extended today, because motivation also affects researchers and managers, and quick success is today one of the most important sources of motivation for these people. Academics and consulting companies look anxiously for quick success and offer a multitude of nearly miraculous theories, techniques and methods that overflow the academic and professional literature, and they frequently forget the philosophies.

If managers look for quick results in quality, it is not infrequent that they forget the rigorous experimentation and the patient search of improvements, and the routine application of Taylorist paradigm could increase the gap between social and technical parts of the system. In the absence of a visible paradigm that supports the techniques and when the philosophies fail, the techniques inevitably do not work well, the improvements of performance are not sustained and the consequences of this kind of actuation are difficult to foresee.