This thesis has its focus within Flexible Automatic Assembly (FAA) systems and its implementation in small companies. The small companies are an important part of a country’s industry and economy. The need for the small company to increase the technology level, in order to be competitive in a global market as a supplier, is evident. The prevailing competence and economical resources of most small companies are often limited, a fact which underlines the need for economical, technological and application flexibility for their assembly system solutions. Therefore, these needs could be viewed as the main objectives posed on solutions which claim to be flexible, such as FAA systems.
The objectives for this thesis are:
- to investigate the state-of-the-art in FAA systems; the primary focus of the investigation residing upon flexibility issues. Theoretical study.
- to investigate the implementation process in a small company that tries to apply FAA technology, and observe the problems that must be solved. Empirical study.
- propose an approach as to how FAA systems should be structured.
As a prerequisite for the proposed structure, the state of the art within FAA systems is discussed with focus on flexibility. The work in this thesis is based on the hypothesis that if systems are developed with properties satisfying the SME needs, the FAA-systems would be user-friendlier both in an economical- and a competence perspective. Hence the need to evaluate the status of current FAA systems.
The work attempts to illustrate, through the theoretical and empirical studies, that there are three main factors that limit success when implementing FAA systems in small companies: Competence acquisition, Product design and Strategy for implementation. Learning, or competence acquisition, is most often carried out in parallel with the actual use of the system. This extensive learning phase most often leads to the system becoming obsolete during its expected run-time (period of use). The product design is another area of interest as it often seems hard to motivate product changes to facilitate automatic assembly, especially if no system is at hand as a driving force. The work also showed that there is a lack of strategy for implementation of FAA. The understanding of the FAA system impact on the product development and manufacturing is not understood, thus, the implementation is treated as a local phenomenon in the assembly shop. Furthermore, the ii uncertainty concerning future development of products is reflected in short payoff times, which makes the economical justification of systems difficult.
All of these factors have been misrepresented to date due to a loose application of the term flexibility, and its implications. Therefore, the work attempts to clarify the flexibility issues, in terms of the assembly process demands, and also proposes a new flexibility map.
To facilitate implementation of FAA into small companies, a modular assembly system approach is proposed. The modules should be task oriented. By using a modular approach, the possibility to have a system that is easy to keep up to date with product development, is greatly improved and the risk to invest in dead ends will decrease. Furthermore, it should become simpler to find an optimal level of automation or to prioritise the flexibility efforts, and thereby reduce the cost for maintaining flexibility over the system lifespan.
A way to develop a decision.making tool for prioritising the flexibility efforts is proposed by using a Penalty-of-Change curve, which exploits the proposed flexibility map as a base. The modular approach is a way to decrease the negative correlation between dynamic flexibility (on-line) and static flexibility (off-line), since too much dynamic flexibility otherwise tends to limit the possibility to change the system to new situations. The modular approach isolates this negative correlation within modules. The technological adaptability will increase if a modular system is used since the standardised interfaces between modules define clear borders between functions or tasks within a system. Furthermore, a modular approach facilitates stepwise knowledge acquisition since it will be natural to focus on modules instead of entire systems.
A leading idea in this thesis is, that small and large companies have the same system demands at module level, which makes it possible to increase the market for standardised modules and thereby decrease the cost for such modules. In other words, standardised modules, instead of systems. Modular systems are not optimised in terms of a low cost/capacity ratio for a certain product generation, instead, the system is optimised towards, reconfigurability to be able to follow product evolution, or totally rebuild for new products. Which, in the long run, may lead to a low cost/capacity ratio. Therefore, a prerequisite for the use of modular systems is that the systems will operate in a frequently changing market where fast ramp-up are more important (to maintain the market share) than an optimised system for each product generation.