Some rules of the brainstorming process are given in the Section Fundamental Design Principles.
Brainstorming is based on an informal approach to problem solving, encouraging people to generate a vast variety of ideas and solutions, sometimes seemingly weird and odd, but that can contain rational seeds and stimulate mental capabilities of people leading them to the expedient results.
In the conventional brainstorming concept, individual brainstorming is considered if not impossible but at least inefficient, since “normal” number of participants should count between six and ten. On the other hand, in most cases group brainstorming can be less efficient because people pay so much attention to other people that they do not generate ideas of their own, or they forget these ideas while they wait for their turn to speak.
Of course, these statements can be contested. If the group does not follow the rules of brainstorming, then such situation is not a problem of the brainstorming concept, but rather of the poor organization of the process. Moreover, when some of the participants do not generate ideas or forget them during discussion, others still do this, and the number of generated ideas by several members of the group in any way can be bigger than the number of ideas generated only by a single person. Contrariwise, one person does not have to worry about other people's opinions and can feel freer and more creative. However, a single person does not have the full set of wide expertise and skills that other team members can add to the discussion.
Therefore, the most widely spread method of brainstorming is the group brainstorming.
The main method of the brainstorming requires involvement of several members of the project team, ideally between six and ten, as it has been already mentioned. This number involves all expertise, imagination, and open mind of all team members. An idea proposed by one team member can be grasped by another one, possibly, more creative and experienced, and taken to the next stage. Thus, group brainstorming is more efficient than individual brainstorming, especially if the participants add different areas of expertise to the process.
Another benefit of group brainstorming is the involvement of all team members and making them to feel the importance of their contribution to the solution, which in turn helps in team building and in development of the team spirit.
One of the popular brainstorming methods is brainwriting, also called “Method 6-3-5”. This method requires six participants; each of them should write down three ideas within five minutes. Then those ideas are handed over to next person which should suggest three similar, and so on. Thus, without any verbal communication the team gets multiple ideas for further analysis, evaluation, and selection of the best fit.
Brainstorming must follow its own rules, the most important of which are:
- A brainstorming session requires a group leader whose role is to keep the ideas as free flowing as possible and to promote creative thinking by encouraging each team member to speak up with anything he or she want to contribute. While admitting all ideas, the leader should keep the discussion on target.
- During the meeting, there is no need to immediately get everyone involved, it is better to give people some time to think on their own. However, it is worthy to let the team members come up with their own ideas in advance, prior to the brainstorming session. They should have some time to ponder over the task and generate some ideas independently, before presenting them with the group.
- In the session, all participants can talk openly, without limitations, but only one conversation at a time should be allowed. The session leader must ensure that participants who usually tend to dominate the conversations would be quietened for the purpose of giving the calm ones a chance to speak up.
- Participants should be encouraged to build on the ideas of others.
- Quality of the proposed ideas is not important; quantity of the ideas is the main goal of the brainstorming session. There is no such thing as a bad idea, and all ideas must be recorded without exception, not only the seemingly good ones.
- Duration of the brainstorming session must be predetermined but flexible.
- Visualization is important during the session, e.g., by using colored markers and Post-its and by sticking ideas on the wall or blackboard for other team members to familiarize with them.
- Someone from the team should be assigned to record all ideas. This allows participants to review afterwards the notes being written in real time and gives everyone else an opportunity to add in something that might have been left out.
- After the end of a brainstorming session enough time should be given to add in ideas, however a clear and distinct time limit must be set up for that.
Selection of evaluation criteria
Selection of evaluation criteria is an important step in evaluation of different concepts developed in the project. Evaluation criteria directly relate to customer’s needs, determining the quality of the proposed solutions and their compliance with the project requirements.
All evaluation criteria fall under five major categories.
In this category all criteria represent the extent to which the objectives of a design project are consistent with customer’s needs, requirement, specifications and goals. E.g., in a car transmission design, how the weight of the gear box meets the weight requirements of the car manufacturer.
These criteria connect available and assigned resources and inputs (funds, expertise, time, etc.) with obtained results. Manufacturing cost of the designed product is an example of such a criterion.
Criteria of this category measure the extent to which the project objectives were achieved, or are expected to be achieved, considering their relative importance. Such criteria, for example, are important in the development of precise technological equipment, such as industrial robots. For the manipulators two major characteristics are the positioning accuracy and the repeatability of the end-effector. These parameters depend on the kinematical and structural characteristics of robot links and joints, as well as on the accuracy and dynamics of the servo drives, so design and selection of these components must correlate with corresponding criteria of effectiveness.
Criteria of impacts help to evaluate long-term effects produced by a development project and/or a designed product. Such impact can be positive and negative, primary and secondary, direct or indirect, intended or unintended. For example, automation of a production line may have both economical (increased production rate and reduced product price on the market) and social (elimination of hard and unskilled labour, creation of new qualified jobs) impacts.
Criteria of this category evaluate the probability of long-term benefits after the project has been completed and the resilience to various risks over time. As an example, such criteria are essential during development of a new bridge construction, when designer must estimate not only economical benefits and impacts in a foreseeable future, but also potential influence of the weather conditions on the structure and its lifecycle.
Generation of alternatives
Given that an iterative process is required to solve an engineering problem and that design constraints need to be respected, generation of the concept alternatives and selection of conceptual design is the most critical step in product development. Concept generation involves multiple steps, as well as several selection methods.
The overall concept generation process consists of the two main phases – ideation and concept screening – and can be accomplished in four major steps.
Step 1. Clarification and decomposition of the problem
Before starting to generate any solution, initial research on the problem and collecting background information is a must. In this step, decomposition of the problem into sub-problems helps to better understand the task and to build a background for the better understanding of the customer’s needs. While decomposing the task into sub-tasks (and the system into sub-systems), you can both look at the big picture identifying potential week links of the entire system, and clarify which characteristic, functionality, and requirements are the most critical, and which ones are needless and can be eliminated.
Step 2. Search for Solutions
Search for solutions is accomplished in two ways: by means of external and internal search.
External search is a part of the information gathering process. Here the team tries to find and then analyze the concepts already existing elsewhere that are related to both the overall problem and/or to the sub-problems. Since the use of existing solutions available on the market can be much easier, cheaper and faster, you should decide whether this approach solves the problem (or any of sub-problems) directly, can solve it with modifications and optimization, or requires a full-scape development process targeting a completely novel solution.
Internal search for solution requires creative and innovative thinking from the design team for the purpose of finding new solutions of the problem. This search is based on the brainstorming session followed with evaluation of various ideas and creation of a short-list of possible solutions.
Step 3. Analysis and Exploration of the Solutions
This step takes up and carries on with the ideas suggested in the previous step. You must be selective about which concepts to pursue from the bunch of those generated during the ideation phase, and which ones to reject. This, on one part, is very important since it may save a lot of time and other resources in the next phases of the design process, and on another part, must be accomplished with caution in order not to overlook potential winning solutions. Concept screening involves organizing and analyzing all the ideas by all possible means including computer simulation and engineering analysis.
The outcome of this step is selection of just few solutions and concepts that are the most promising and beneficial for further consideration.
Step 4. Refining the Solutions
The entire design process is an iterative process. The same applies to any of its phases and steps. Therefore, the concept generation process is also cyclical. With any new information and conditions popping up during concept generation and evaluation, you may need to go back into a previous step, whether it is a brainstorming session, or even a data search. In this way, the list of proposed solutions can be refined and improved.
Engineering analysis and mathematical modelling in design
Types of engineering analysis
Most of the design activities require different types of engineering analysis that address specific aspects of the design process and the product under design. Engineering analysis can be applied to any type of product development and accompanying research.
Roughly, the types of engineering analysis can be divided into two main categories: qualitative and quantitative analysis. However, such classification is very relative, since each of these categories may require involvement of the other one during design process.
The main qualitative types of the engineering analysis are:
- Functional analysis. This type of engineering analysis considers the functionality of the new product. It takes place during the conceptual design phase, allowing you to define whether the product should have one or more functions, and to specify which one can and should be added to the proposed design. Depending on the results of the functional analysis, various solutions and possible product configurations are proposed on the conceptual design phased for further evaluation and selection;
- Value analysis. The value engineering analysis targets the consumer properties of the goods. During this analysis, you can evaluate the possibilities and relevance of adding to the product extra functional capabilities and characteristics bearing high value and quality in comparison to the competition already existing on the market;
- Safety analysis. This type of analysis must be accomplished in every single step of the product development process and address current norms and regulations on human safety and health, as well as impact on the society and environment. The main consideration here is whether the new product presents any direct or indirect risk either to individuals or to general public, and the goal is to assure no harm is caused at all. Such analysis often calls for simulation to evaluate possible damage to property or to the environment and helps manufacturer to avoid liability or possible financial responsibility for any injury or damage in consequence of a poor and unsafe design;
- Ergonomic analysis. Ergonomics as well involves health and safety issues; besides, it considers human comfort and convenience during operation. Ergonomic analysis also takes place during conceptional design of the new product, with people and their interaction with the latter becoming an important factor.
Quantitative engineering analysis includes various types of mechanical and structural analyses. The most common mechanical and structural analyses include:
- Linear and non-linear structural analysis considering behavior of a structure under specific conditions;
- Fatigue analysis determining the endurance of the product;
- Vibrations analysis monitoring the levels and patterns of vibrations within a part, machine or structure, for detecting abnormal vibrations and determining the methods of preventing shocks and excessive vibrations leading to object’s failure;
- Heat transfer analysis considering capability of the design to keep or dissipate the heat generated by moving components (e.g., in electrical motors) or obtained from an internal or external heat sources (e.g., in a passenger bus equipped with a heating/cooling system).
Mechanical engineering analysis can be performed for both the entire design and for each of its subsystems, helping to evaluate and compare different alternatives, and decide which one better meets the requirements.
Quantitative engineering analysis is based on mathematical models that include several phases of the development. When engineering analysis is accomplished, the obtained parametric (mathematical) solution is transferred to the physical model by converting and implementing the obtained results into a real-time engineering model.
Main stages of general engineering analysis
- Identification of the physical problem and forming the list of specifications for it. This stage coincides with the formulation of the customer’s needs and problem definition.
- Idealization and assumptions of actual physical situations for subsequent mathematical analysis for geometry, loading conditions, constrains and boundary conditions.
- Mathematical modeling and analysis using suitable mathematical formulations/model and obtaining solution of a specific engineering problem.
- Interpretation of results: conversion of the analytical results into physical parameters and requirements, with following implementation of the latter into engineering solution.
Tools for engineering analysis
- Finite Element Analysis (FEA) - the main method in design analysis to simulate physical behavior of a designed product. The FEA process consists of subdividing all systems into individual elements whose behavior is easily understood and described and then rebuilding the original system from these elements.
- Structural Simulation, that is based on FEA and allows to predict part’s behaviors and solve structural problems during design process. It can be used for simple tasks such as a dynamic strain analysis of a single part, or for complicated statically indeterminate structural analysis composed of multiple parts, connections, material nonlinearity, and contact regions.
- Stress Analysis - the most common type of design analysis in the field of mechanical engineering. Stresses, both structural and thermal, help to determine reliability of a part; using stress analysis engineer can evaluate whether the part will fail and in what specific region, and whether design modifications are necessary to resolve the problems.
- Thermal Simulation, essential in different engineering applications, especially in electronics where the power of the elements constantly grows while packaging becomes tight and miniscule. This leads to the higher heat generation in the reduced space volume. Another application is lighting and heating devices which also need to meet thermal performance requirements.
- Kinematic and Dynamic Analysis, applied to the products with moving parts, where parts’ displacements, velocities, and accelerations, as well as the forces they generate or are subject to, are essential for the product behavior and functionality.
Evaluation of design solutions and decision making
When all customer needs and designed requirements are fully defined, you can start to generate multiple design solutions, or concepts. This procedure was described in detail earlier. The next step is to choose only one solution that matches the initial requirements and specifications in a best way.
When you need only one concept for the implementation, you can make your choice between the proposed solutions in two different ways. The simplest way is when one of the solutions fully meets all requirements of the initial task. It happens mostly with simple systems that does not have any subsystems or has a very limited number of them. Such system architecture makes it easy for you to propose different concepts, because each proposed solution fully addresses all requirements. In this case you simply need to compare the concepts’ properties and select the most prospective design, using evaluation criteria.
However, often happens that the system includes several subsystems. If so, the comparison and evaluation must be accomplished not on the system level but rather on the subsystem level, meaning that the corresponding subsystems are evaluated and compared between each other, and the final system is then built from the winners of the subsystem competition. In this situation the low-level subsystems are evaluated, but the top-level system may not require further evaluation (although it is not always the case, since sometimes combination of the subsystems may also affect the performance of the system and this issue has to be addressed as well).
The principles of selecting evaluation criteria were presented in the previous sections. Evaluation criteria may be used as well as optimization criteria when optimization is parts of the design process. Here an example of a practical approach to the selection process is given.
Let us consider a project where a special gear train for a vehicle must be designed. After making a research on the best practices in the industry, the design team compiled a list of 19 possible criteria that can be used for the concept evaluation. The client was asked to assign weights to all criteria between 0 to 10, where 0 means the lowest weight of criteria and 10 is the highest weight. As it can be seen from the table, the highest weight was assigned to eight criteria from the initial list.
Eight criteria are also too much for the design process, thus the team reduced the list further by choosing only those criteria that have the maximum weight 10. The new list was presented to the client with the request to rank it with the letters, where “A” means the highest rank, “B” is lower, and so on.
After the second ranking the team finalized the list of the design criteria with five of them having the ranks from “A” to “D”.
Returning to the selection of the concept to develop, you may use an evaluation procedure consisting of the three major steps:
- Screening, that takes place after the brainstorming session. Here, the team must eliminate all those concepts that are not feasible, too extravagant or fall out of the norms and practices adopted in the industry.
- Comparison, where all remaining concepts must be evaluated and compared in compliance with the adopted evaluation criteria; advantages and disadvantages of each concept must be considered.
- Decision making, and here the best concept is selected for further development. The “best” here means the concept that obtains the highest rank after analysis of all concepts with respect to the chosen evaluation criteria. It may also happen, that the winning concept is generated as a combination of different components taken from different concepts.
One of the most effective tools for the selection among several candidate design options is the Pugh matrix. The matrix is built when evaluation criteria are identified and clearly defined.
One of the candidate design options is used as a baseline. As a baseline designer can also use a solution which already exists (on the market in general, or in the production facility of the customer), if the project targets developing of an alternative solution to the existing one. In any case, in the matrix the proposed solutions are compared with the baseline by each of the selected criterion.
When each candidate design option is compared criteria by criteria (or requirement by requirement) against the baseline design, the result is put down into the table as a pairwise score with the symbols: “S” (same), “+” (better) and “–“ (worse). Some people prefer to use numbers instead of the symbols: 0, +1 and -1, correspondingly.
It is also possible to add extra levels of distinction by using “++” (much better) and “– –" (much worse). Another option could be using the series of numbers 1 to 5 where the “same” is denoted as 3; 1 and 2 being “much worse” and “worse”, correspondingly; 4 and 5 being “better” and “much better”.
In the next step, for each concept (design option) the total score is calculated by summing the numbers of “+”, “S” and “–“. The highest ranked score formally is the “winner” but it should not be taken for granted, and personal common sense should be also used for judgement.
To facilitate the final selection, the weights of each criteria can be added to the table according to their importance for the selection process. In this case, during summing each symbol has its own factor equal to the corresponding criteria weight. The weights are especially effective and useful when “pure” summing of the ranks does not reveal an obvious winner. This method is called the Weighted Objectives Method, and it allows the decision-maker to consider the difference in importance between criteria.
Final design solution can be also built from a combination of the “best” components of different alternatives, i.e., the choice will be done in favor of a “hybrid” design solution.