Observations on Pahl and Beitz Design for Manufacturing Methodology
Pahl and Beitz start Engineering Design: A Systematic Approach by discussing the historical context under which systematic design was developed. They discuss the fundamental aspects of the systematic approach. Pahl and Beitz then move into product planning methods useful to designers. Finally, Pahl and Beitz move into project workflow management [1].
Pahl and Beitz have determined that design is a highly psychological process where consideration must be made for the individuals involved in the process as well as the technical details of the project. Using the systematic design processes that Pahl and Beitz outline, design optimization is possible in a way that minimizes conflicts between mutually exclusive constraints.
In an organization, there are multiple methods of splitting departments. The two primary styles are product-oriented organizations and problem-oriented organizations. In a product-oriented organization, all aspects of a specific product type are maintained by a specific organization. This style of organization is also called commodity-based. In this style, one organization may be responsible for all aspects of one particular widget.
Contrasting to that style of organization are problem-oriented organizations. These organizations are responsible for one particular function of the engineering lifecycle. For example, one organization may be responsible entirely for stress analysis. This organization performs stress analysis on all products. These organizations are also called functional-type organizations.
Before considering the particulars of a design methodology, it is important to consider the objective of a design methodology. Design methodologies must incorporate a problem-directed approach to solving functional requirements. Design methodologies should not be constrained to a particular field or discipline. Design methodologies must seek optimal solutions while still allowing for a robust management system to exist simultaneously. In addition, design methodologies must not rely on strictly intuition or chance.
Adequate solutions must meet five standard criteria. Production costs must be minimized by the solution. Space requirements should be minimized whenever feasible. Similarly, weight must also be minimized. The manufacturing processes must involve minimum losses. The parts and assemblies should invoke methods such as Design for Manufacturing to optimize handling.
Paul and Beitz discuss various design methodologies throughout the text. The different methodologies of note are adaptive design, variant design, conceptual design, embodiment design, detail design, and computational design. Each specific method is ideally used under certain conditions and at certain stages in the overall project.
Conceptual design is the design methodology by which solution principles are not yet known. This design stage seeks to understand the problem statement and establish a solid solution space for the given problem and associated tasks. Conceptual design seeks to create a general solution for the specified problem. It does not seek to create an exact solution
Embodiment design is the stage of design following conceptual design. This design methodology seeks to constrain the solutions presented in the conceptual stage in order to establish a firm design. Embodiment design involves the creation of measurements and material selection.
Detail design is the process by which a product is finalized. Production methods are determined during this stage. Operational characteristics are calculated and published. This stage will only involve minor redesign as most aspects of the product have already been finalized.
Computational design is the final stage of the design process. This stage allows for the collection and elaboration of all performance characteristics of the product. Final stress analyses will be conducted during this stage. Any necessary certifications will be performed immediately following this stage, the product will be considered complete.
Adaptive design is the process by which an embodiment design utilizes existing solutions to reach a conclusion. Adaptive design makes use of standard solutions to standard problems. Design catalogs may be used to further expedite the design process.
Variant design is similar to adaptive design; however, it varies in that only the size and geometric arrangement of previous solutions are changed. This allows for rapid solution creation while also reducing the time and energy required to reach a conclusion. Only dimensions and mating are changed between solutions. Variant design may also be referred to as principle design.
Pahl and Beitz also discuss the nature of functions in the text. Functions are input/output relationships that accomplish the completion of a task. Functions may be either the overall function for a task or a designated subfunction that combines with other subfunctions to complete a task or be utilized individually to complete one subtask of a given task. Functions typically accomplish tasks such as acting as part of a feedback circuit or performing torque transfer between systems.
Logical relationships can be created from task-specific functions. The if-then relationship that follows from designs may be reconsidered in the context of logical flow diagrams. These diagrams use standard Boolean logic to create flowcharts that describe the functions of a system. AND gates specify when all functions must be completed to achieve a result. OR gates specify when one function of a set must be accomplished to achieve a result. Conversely, NAND gates and NOR gates specify the opposite of their juxtaposed gate. Exclusionary gates may also be used though Pahl and Beitz did not discuss this in the text.
Functions may also be discussed in the context of their impact on the system. Functions may manifest in the form of input effects from the user’s decisions that change the system. System changes may result in feedback effects being created that result in intended or unintended system operations. Disturbing effects are undesirable results that occur due to influences from external forces. Disturbing effects impede functions from carrying out their intended purpose and are generally intentional in nature. Side effects are similar to disturbing effects; however, they are unintended in nature [1].
In general, I believe the assertions of the authors. Each defined design methodology is both logical and well thought out. Despite being unaware of the taxonomy and nomenclature of these methodologies, I have been using them professionally since I was an undergraduate. The descriptions provided of functional design, though lengthy, is well thought out and fully compatible with how I was trained to accomplish system design.
In terms of missing material, the only thing I noticed was that the logical flow section could have been improved. Though nothing was incorrect, I feel that greater discussion could have been made by the authors to discuss other use cases of the function logic system. Logic flowcharts are used extensively in electrical and software engineering to map out desired states. Similarly, I was surprised that there was no discussion of state machines.
State machine representations, such as the Moore state graph, are used to demonstrate function flow between machine states that are unique to a particular sequence. Using a strictly logical flow system does not allow for time-dependent or variable input functions to be accounted for. An example may be considered in the form of a simple adder circuit. The most significant bit used is based entirely on the history of signals that have already been seen. While a purely logical flow may only account for one iteration of the adder circuit’s function, a Moore state graph accounts for each subsequent iteration throughout the expected range of the system [2].
I intend to use this material to further enhance my design methodologies used in industry. The material provides a formal codification of many principles I have already been practicing. Using a formalized process will allow for a more normalized workflow as I progress from project to project. I will most likely print out several of the figures from this section and keep them at hand to reference when performing future design work.
One unexpected bit of information I plan on using in an unconventional way is the discussion of organizational arrangements in the first chapter. I recently was invited to provide inputs for the reorganization of my department and our future-state structure. I will likely invite my management to consider the arrangements suggested by Pahl and Beitz.
I do not consider the material to be too rigid. However, I can see that some designers may consider it to be such. As I have worked in the aerospace and nuclear industries, I am procedure-driven by nature. I see formal codification of design rules and methods as a positive influence. I would imagine that individuals from a less strict background may find the codification of design principles to be restrictive
References:
[2] Roth, C., Kinney, L., 2014, Fundamentals of Logic Design 7thEdition, Cengage Learning, Stamford, CT.
[3] Jensen, J., 2006, A User’s Guide to Engineering, Pearson, Saddle River, NJ.
[4] Friedenthal, S., Moore, A., Steiner, R., 2012, A Practical Guide to SysML: The Systems Modeling Language, Elsevier, Waltham, MA.
[5] Bai, H., Arcak, M., Wen, J., 2011, Cooperative Control Design: A Systematic, Passivity-based Approach, Springer, NY.
[6] Eder, W., Hosnedl, S., 2010, Introduction to Design Engineering, CRC Press, Boca Raton.
[7] Farid, A., Suh, P., 2016, Axiomatic Design in Large Systems: Complex Products, Buildings and Manufacturing Systems, Springer, Switzerland.
Member discussion