The Life Cycle of Manufacturing Networks in the Mass Customisation Era

Evolution of manufacturing and current challenges

Evolution of manufacturing paradigms

Over time, manufacturing paradigms, driven by the pressure of the environment in which they operate, change in character and evolve in patterns (Fig. 2). The various patterns witnessed up to now can be roughly correlated to movements between three stages: (1) craft shops that employ skilled artisans, (2) long-linked industrial systems using rigid automation, and (3) post-industrial enterprises characterised by flexible resources and information intensive intellectual work. Prevailing manufacturing paradigms are, in chronological order of appearance, the following: craft production, American production, mass production, lean production, mass customisation, and global manufacturing. Apart from American production, all other paradigms are still "operational" today in different industrial sectors.

Fig. 2 Evolution of manufacturing paradigms (adapted from [11])

By studying these notable transitions, which are attributed to the pressure applied by social needs, political factors, and advances in technology, it is noticeable that factory systems and technologies have been evolving in two directions. Firstly, they increased the versatility of the allowable products' variety that they produced. This resulted in numerous production innovations, design technology advances, and evolution in management techniques. Secondly, companies have extended factories like tools and techniques. Factories emerged from firms that introduced a series of product and process innovations that made possible the efficient replication of a limited number of designs in massive quantities. This tactic is widely known as economies of scale. Factory systems replaced craft modes of production as firms learned how to rationalise and product designs as well as standardise production itself. Although factory organisations provided higher worker and capital productivity, their structure made it difficult to introduce new products or processes quickly and economically, or to meet the demands of customers with distinctive tastes; factory-oriented design and production systems have never completely replaced craftsmanship or job shops even if the new technologies continue to appear. The result, in economic, manufacturing, and design concepts, has been a shift from simple economies of scale, as in the conventional MP of a limited number of products, to economies of scope and customer integration. It is clear that MP factories or their analogues are not appropriate for all types of products or competitive strategies. Moreover, they have traditionally worked best for limited numbers of variants suited to mass replication and mass consumption. The craft approach offers a less efficient process, at least for commodity products, but remains necessary for technologies that are still new or emerging and continues to serve specific market niches, such as for tailoring products for individual needs and luxury or traditional items. A categorisation of the different production concepts based on the indicators system reconfigurability, demand volatility, and product complexity is depicted in Fig. 3.

Fig. 3 Characterisation of production paradigms based on demand structure, product complexity, and product flexibility

Today, issues introduced by the shift of business models towards online purchasing and customisation must be tackled in cost-efficient and sustainable ways in order for companies to maintain their competitiveness and create value. To respond to consumer demand for higher product variety, manufacturers started to offer increased numbers of product "options" or variants of their standard product. Therefore, practice nowadays focuses on strategies and methods for managing product, process, and production systems development that are capable of supporting product variety, adaptability, and leanness, built upon the paradigms of MC and product personalisation. The currently widespread MC is defined as a paradigm for "developing, producing, marketing and delivering affordable goods, and services with enough variety and customisation that nearly everyone finds exactly what they want". This is achieved mostly through modularised product/service design, flexible processes, and integration between supply chain members. MC targets economies of scope through market segmentation, by designing variants according to a product family architecture and allowing customers to choose between design combinations. At the same time, however, MC must achieve economies of scale, in a degree compared to that of MP, due to the fact that it addresses a mass market. Another significant objective for companies operating in an MC landscape is the achievement of economies of customer integration in order to produce designs that the customers really want. On the other hand, personalised production aims to please individual customer needs through the direct integration of the customer in the design of products. The major differences between the prominent paradigms of MP, MC, and personalisation in terms of goals, customer involvement, production system, and product structure are depicted in Fig. 4.

Fig. 4 Differences between production paradigms (adapted from [20])

A research conducted in the UK related to automotive products revealed that 61 % of the customers wanted their vehicle to be delivered within 14 days, whereas consumers from North America responded that they could wait no longer than 3 weeks for their car, even if it is custom built. Such studies point out the importance of responsiveness and pro-activeness for manufacturers in product and production design.

During the last 15 years, the number of online purchases has increased and recent surveys show that 89 % of the buyers prefer shopping online to in-store shopping. Web-based and e-commerce systems have been implemented and have proved to be very effective in capturing the pulse of the market. These web-based toolkits aim at providing a set of user-friendly design tools that allow trial-and-error experimentation processes and deliver immediate simulated feedback on the outcome of design ideas. Once a satisfactory design is found, the product specifications can be transferred into the firm's production system and the custom product is subsequently produced and delivered to the customer. Still online 2D and 3D configurators do not solve practical issues such as the assembly process of these unique variants. Although proposed approaches include e-assembly systems for collaborative assembly representation and web-based collaboration systems, the research in this area needs to be expanded in order to provide tools for assembly representation and product variant customisation. An additional constraint is that globalised design and manufacturing often require the variants for local markets to be generated by regional design teams, which use different assembly software and source parts from different supply bases. The incorporation of the customers' unique tastes in the product design phase is a fairly new approach to the established ways of achieving product variety and entails significant reorganisation, reconfiguration, and adaptation efforts for the company's production system. Variety is normally realised at different stages of a product life cycle. It can be realised during design, assembly, at the stage of sales and distribution, and through adjustments at the usage phase. Moreover, variety can be realised during the fabrication process, e.g. through rapid prototyping.

It should finally be noted that naturally, even if the trends dictate a shift towards personalised product requirements, it should always be considered that forms of production such as MP cannot be abandoned for commodities and general-purpose products, raw materials, and equipment. After all, paradigms are shaped to serve specific market and economical situations.

Globalisation

Globalisation in manufacturing activities, apart from its apparent advantages, introduces a set of challenges. On the one hand, a globalised market offers opportunities for expanding the sphere of influence of a company, by widening its customer base and production capacity. Information and communication technologies (ICT) and the Internet have played a significant role to that. On the other hand, regional particularities greatly complicate the transportation logistics and the identification of optimum product volume procurement, among other. Indicatively, the difficulty in forecasting product demand was highlighted as early as in 1986 by the following observation from Intel laboratories: when investigating the match between actual call off and the actual forecast, they estimated that supply and demand were in equilibrium for only 35 min in the period between 1976 and 1986. Enterprises started locating their main production facilities in countries with favourable legislation and low cost of human labour; thus, the management of the supply chain became extremely complex, owing primarily to the fact that a great number of business partners have to mutually cooperate in order to carry out a project, while being driven by opportunistic behaviours. Thus, manufacturing networks need to properly coordinate, collaborate, and communicate in order to survive.

On a manufacturing facility level, the impact of supply chain uncertainties and market fluctuations is also considerable. The design and engineering analysis of a complex manufacturing system is a devious task, and the operation of the systems becomes even harder when flexibility and reconfigurability parameters must be incorporated. The process is iterative and can be separated into smaller tasks of manageable complexity. Resource requirements, resource layout, material flow, and capacity planning are some of these tasks, which even after decomposition and relaxation remain challenging. In particular, in the context of production for MC businesses, issues such as task-sequence-dependent inter-task times between product families are usually ignored, leading to inexact, and in many cases non-feasible, planning and scheduling. Even rebalancing strategies for serial lines with no other interdependencies is challenging, leaving ample room for improvement in order for the inconsistencies between process planning and line balancing to be minimised.

From a technological perspective, the increased penetration of ICT in all aspects of product and production life cycles enables a ubiquitous environment for the acquisition, processing, and distribution of information, which is especially beneficial for a globalised paradigm. With the introduction of concepts like cyber physical systems (CPS) and Internet of things (IoT) in manufacturing, new horizons are presented for improving awareness, diagnosis, prognosis, and control. Also, the relatively new paradigm of agent-based computation provides great potential for realising desirable characteristics in production, such as autonomy, responsiveness, distributiveness, and openness.