Integral Logistics Management — Operations Management and Supply Chain Management Within and Across Companies

7.4 Generative and Adaptive Techniques for Engineer-to-Order

Intended learning outcomes: Differentiate between the classical procedure and different archetypes of engineer-to-order. Describe the approach for basic and for repeatable engineer-to-order.


Although mass customization products are nearly all physically different, many companies that consider mass customization to be one of their core competen­cies view such products as “standard” products. This is because they can all be produced using a make-to-order process. For such companies, “non-standard” or “customi­zed” are terms that refer to products that cannot be described with the configurator, and thus need an engineer-to-order environment. According to Figure 4.4.5.2, the engineer-to-order production environment is strongly related to products according to (changing) customer specification. According to Figure 4.5.2.1, nonrepeti­tive or “one-of-a-kind” production, often together with mass customization, are the corresponding production types. The customers’ requi­rements here, in terms of speed and cost, are not as stringent as for “standard” (mass customized) products, but in many branches they are getting closer to mass customization requirements. Also compare Figure 7.1.2.1.


7.4.1 Classical Procedure and Different Archetypes of Engineer-to-Order

In the plant manufacturing industry, many areas of a plant facility are customer specific and produced in nonrepetitive production. With an intelligent product concept, however, it is usually possible to determine similarities of the plant facility to previously produced plants. For instance, during processing of an order, the vendor recalls previous “similar” problems. Derivation can thus often be performed on the basis of a previous customer order as a “parent version,” using adaptive techniques according to Section 7.2, at worst position by position. Such orders generally require a high degree of order-specific engineering. They have to be developed and built to exactly match the customer's specifications. Other examples of this classic situation would be customer-specific modernization of an aircraft, an oil platform, or a nuclear power plant. If there are very many bill-of-material and routing-sheet positions, this will entail a high load on qualified employees. In addition, the lead time is long. Either is only justified for high-value-added products.

In the plant manufacturing industry, there have been attempts to restrict order specific engineering to a minimum and to use generative techniques for the larger part of the customer order (also see Section 7.3 below). This has worked well in the exterior construction business; for example, where certain elements of building exteriors are selected from a preset range of variants. Combining the elements for the whole face of a building, however, may well require order specific engineering. Except in the area of project management, or general procedures, this minimum level of ETO offers little or no potential for standardization and automation. The consequence is a low degree of industrialization. As well as this classic case of engineer-to-order, there are also situations where a higher degree of industrialization is necessary (and also possible) for reasons of cost or throughput time. The Portfolio in Figure 7.4.1.1 shows different archetypes of engineer-to-order (ETO).

Figure 7.4.1.1  Different archetypes in engineer-to-order (compare [WiPo15]).

The “Annual units sold” dimension shows how many units of a product family are sold on average per year. A unit is generally the lot that is ordered as one order unit and therefore manufactured in this lot size. Often, the lot size is one; then, the units sold are the pieces sold. The threshold of 750 units/year is based on the assumption that making at least two units/day offers a sufficiently high degree of repetition for cost-effective automation.

The “Engineering Complexity” dimension measures the number of necessary engineering hours per unit sold. In practice, order-specific engineering is one of the biggest drivers of complexity in the ETO environment. Consequently, the possible benefit of an automation depends on it. The boundary between the quadrants has been set at 2000 engineering hours/unit, as that corresponds to one person year.

The classic “engineer-to-order” situation described above is called “Complex ETO” in Figure 7.4.1.1. Only a small degree of industrialization is possible, since the repeat rate is very low at both the product and the process levels. “Non-competitive ETO” relates to firms that deploy over 750 full-time positions for customer-specific engineering (≥ 750 units/year times ≥ 2000 h per unit). For firms of this size, it was previously possible to combine them as a conglomerate of subsidiaries, with each company supplying less than 750 order items per year. The subsidiaries can therefore be included in the “Complex ETO.” We suspect that it is not profitable over the long term for a company to position itself in this quadrant, unless it has a monopoly and the company is therefore not exposed to competition. The two others, “Basic ETO” and “Repeatable ETO,” are covered in the next two sections.


7.4.2 Approach for Basic Engineer-to-Order

The lower the value added to the order, the lower the effort that should be expended on customer-specific adaptations. Hence the term “Basic”engineer-to-order for these arche­types, where engineering complexity needs to be kept to a minimum. Cable cars and asphalt mixers are good examples of this. On the one hand, cable car operators expect a cable car design that is specific to their ski area. On the other hand, the equipment as a whole regularly has to be adapted to the topographical conditions. Typical reasons for customer-specific adaptations to asphalt mixers are regional environmental standards, or harsh climatic conditions. Examples where even less engineer-to-order can be carried out for each order would include fastening tools or air diffusion grilles. But customers often order an entire lot of items that are specifically adapted for a particular construction project.

In order to reduce the engineering complexity, a medium degree of standardization and automation is recommended here. However, the low order repeat rate is a limitation. Standardization is primarily based on commonality of components and a modular product concept. Building on that, a product family is often developed for the production environment on a make-to-order basis and a mass customization production type. If the product and process configurator cannot completely specify the customer order, the provisional configuration result is adapted using adaptive techniques, where the bill-of-material positions and operations can be added, amended, or deleted.

A limited degree of automation is determined by the definition of product families with an unfinished product structure, and the development can then be finalized according to customer requirements. Figure 7.4.2.1 illustrates the technique.

The unfinished product structure looks like a template for a product structure. It contains materials that a company (e.g., a sheet metal working firm) typically uses as a starting point for a product (e.g., various aluminium or steel sheets), and a sequence of loosely descri­bed operations in which the firm has recognized expertise (e.g., cutting, bending, assembly) or that they get done by external suppliers (e.g., surface treatment). Components or operati­ons may also be entered as dummy positions. The customer’s order parameters are used by the configurator as much as possible to decide the positions that have to be part of the order. The configurator stops wherever in the template a “?,” i.e., the symbol for an informa­tion gap, is encountered and asks for entry of the attribute value specific to the current customer order, e.g., the quantity per for the components C1, C2,…, Cn. The result, although inter­mediate, is often already useful for initial cost calculations (for example, where +/- 10% accuracy is sufficient) and for logistical control during order execution. The intermediate result is then updated as necessary using adaptive techniques, according to the data from the customer, by adding, changing, and removing bill of material and routing positions.

This technique allows a certain degree of automation, as the same unfinished product structure can be used for many customer orders. In practice, it has been observed that automation in “Basic” engineer-to-order is primarily achieved through the use of product configurators in the tendering process.

Fig. 7.4.2.1        Template for bill of material and routing sheet used to work out similar variants.


Routing sheet



7.4.3 Approach for Repeatable Engineer-to-Order

This archetype would include elevators or buses, for example. As with the “Basic” engineer-to-order, a product family is often developed for the production environment on a make-to-order basis and a mass customization production type. If necessary, the result of the configuration process is adapted according to the customer order. This approach is naturally only suitable for product families that only need partial customer-specific adaptation.

Since these are often “engineer-to-order” jobs, the level of competition is high. On the other hand, there is often repetition, at least at process level. This fact can and must be used for a high degree of standardization and automation. For example, in the segment of skyscraper elevators, the elevator in the top floor must fit with the architect’s personal concept. The “engineer-to-order” process requirements are thus similar to those of the fashion sector. In addition, selling the “non-standard” (i.e., “engineer-to-order”) elevator on the top floor is often the order winner for the many “standard” (i.e., “make-to-order”) elevators in the whole building. Therefore, the specific engineering of the top floor elevator is not adequately remunerated. Thus, the engineer-to-order process must be fast and efficient.

At the fundamental level, the use and integration of product configurators in the ERP and CAD software systems can help automate the tendering and ordering process from end to end. Expertise and experience with suitable technical methods and tools (e.g., product configurators) are used in the individual sell, engineer, and make processes. Even as early as the sales process, configuration allows for an initial cost calculation and for a virtual product, which allows the customer to experience something as close as possible to the physical product.

Fast and efficient engineer-to-order also imposes high organizational requirements. And that is a key reason why customer-specific adaptations are normally not handled by product design, but as a separate process by a department that specializes in carrying out customer-specific adaptations. Figure 7.4.3.1 shows a typical engineer-to-order business process in a company.

Fig. 7.4.3.1       A typical engineer-to-order business process and its permanent enabling process (compare [Schö12]).

The sales phase (including quoting) is followed by receipt of the customer's order, followed in turn by design, production, and delivery. In practice, fast and efficient engineer-to-order calls for the consistent, long-term use of an enabling process. Figure 7.4.3.1 calls this process a permanent engineer-to-order enabling process. Queries from the ongoing engineer-to-order business process are answered through a form of know-how transfer. If additional know-how is gained during order execution, it is fed back to the enabling process in the form of lessons learnt. At the organization level, this means expertise and experience of dealing with the customer’s engineer-to-order requirements. This in­volves the business models between the company and external customers and suppliers, but it also means the “business model” around the internal customer-supplier relationships between sales, engineering, and production, at all levels. At process level, this means expertise and experience of managing how the company works with external customers during the product specification and manufacturing phases, but also how internal customers work with suppliers. Another point is know-how of user interaction with the product in a virtual status, i.e., before and during the physical manufacturing.

Whilst in the classic mass-customization culture it has been possible over the years to concentrate the mass-customization expertise across fewer people and to focus on the design pro­cess (and, to an extent, on the sales process), a quick and efficient engineer-to-order system needs this expertise shared amongst more people, and it must in turn be extended to the workshop. So we can talk here of a distinct engineer-to-order culture.

For companies that consider mass-customization products as their “standard” products, it is notable that an engineer-to-order (“non-standard”) customer order (all too) often results in a new parameter, which controls the new customer-specific components. The overhead of introducing the new para­meter is only covered if the same customer orders their special variant several times. To solve this problem, we observe that “non-standard” products will use the principles of commonality and of modular product concept that have led to increased productivity in the case of mass customization. Here, these principles can and must also be applied for parameterized component families, because components of product families are sometimes families in their own right. However, their parameters may have a different name or semantic structure than that of the product family, perhaps because the component family already existed beforehand for another product family. Transforming the parameter values from the product family efficiently to those of the component family increases the commonality of the component family. This often helps in avoiding new parameters or components. One more thing: The increased commonality of the component family also encourages product innovation.

Therefore, an important role of the permanent engineer-to-order enabling process is to carefully determine the parameterization for product families, particularly for component families. This task includes (1) determining and maintaining the component families and their set of parameters, (2) increasing re-usability of parameterized component families, thereby ensuring their commonality, and (3) encouraging colleagues to use existing parameters or to suggest sensible enhancements (especially for the value ranges). This task is some­what more difficult than the comparable task of ensuring that single components used for newly constructed products can be re-used. The leaders of the teams that carry out these centrally organized tasks are very experienced product or process developers. They understand the reasons for the existing parameteriza­tion, and can follow the thinking of their colleagues when developing. They also have excellent social skills, to encourage colleagues to support their standardization efforts.



Course sections and their intended learning outcomes

  • Course 7 – The Concept for Product Families and One-of-a-Kind Production

    Intended learning outcomes: Produce logistics characteristics of a product variety concept. Explain adaptive and generative techniques in detail. Describe the use of generative and adaptive techniques for engineer-to-order. Differentiate various ways of cooperation between R&D and Engineering in ETO Companies.

  • 7.1 Logistics Characteristics of a Product Variety Concept

    Intended learning outcomes: Differentiate between high-variety and low-variety manufacturing. Describe different variant-oriented techniques, and the final assembly schedule.

  • 7.2 Adaptive Techniques

    Intended learning outcomes: Explain techniques for standard products with few variants as well as techniques for product families.

  • 7.3 Generative Techniques

    Intended learning outcomes: Disclose the combinatorial aspect and the problem of redundant data. Present variants in bills of material and routing sheets as production rules of a knowledge-based system. Explain the use of production rules in order processing.

  • 7.8 Scenarios and Exercises

    Intended learning outcomes: Apply adaptive techniques for product families. Disclose the use of production rules in order processing. Elaborate the setting the parameters of a product family.

  • 7.9 References

    .


Print Top Down Previous Next