Introduction to Design

We all have designed and produced (assembled) something, whether it was in sand, LEGO, clay or some old wooden boards during childhood to more complex projects during adult life. So do we need to know anything about design since we can do it intuitively? Well, the answer is both yes and no. You don’t have to think of design methods or such when creating single or less complex projects where perhaps you is the only user of it. However if you want to make a design a product that is going to be produced in some numbers the importance of using design and engineering methods increase.

Today a lot of the methods, technologies and software that are used in bigger product development projects can be found in lighter (simpler) versions on the Internet.

On this webpage an introduction to design and product development methodologies is going to be introduced. This is by no means a complete description but at least it will give you some understanding of what to think of, what tools and methods that can be used and some hints on glossary and words that can guide you further in your search for knowledge.

 

Design, Engineering Design and Engineering?

In everyday life the word Design is used differently in some languages. In Scandinavia we are used to speak of ”that’s a nice design” but then we often only refer to the shape, how physically pleasing something is or that something is really designed for it’s purpose. We don’t really mean that the thing is beautifully engineered or that the functionality is excellent.

Design can be seen from many perspectives. In Nordic terms when we speak of Design the formal description is ”Industrial Design”. Which translates ‘formgivning’ in Swedish. Whereas the English term ‘design’ relates to engineering and making stuff work in a certain way, i.e designed for a purpose.

With Engineering Design we mean the way things are engineered, that is, the process of designing stuff with engineering tools and methods. In Swedish ‘Engineering Design’ translates to ”konstruktionsmetodik” . It is important to understand that it’s a process that is in focus here, that is, we use systematic ways to create, design, analyse and evaluate, judge and take decisions about the product and the business.

Engineering on the other hand refers to tools and methods that are used when designing to analyse the designs behaviour and properties in the real world.

There is however no real definition of the word design. And all these terms are heavily related to each other. So let’s skip the history and definitions’ and dive into the fun stuff.

Design as a Process

In this article we are going to talk about design as a process. That is, you have some tasks or things that you want to perform to solve a specific need or to be used for a specific purpose. You wish solve this need by some physical artefact that helps in realizing the tasks and which you can sell as a “product”. I choose to talk about the process cause this is where you realize your product. If you create any product or service you use a process to realize it, whether it is conscious or not.

The purpose of discussing the process is that it is supposed to help you think and work in a methodical way. Some things we can think of right away but some parts may only be revealed if using a process that forces you to think, design and analyse your product from different angles and perspectives.

There exist numerous design processes and views on how a design process should look like and all of them have pro’s and con’s. I’m going to give an excerpt of some of them.

In the past the design process was seen as an isolated thing where you first designed your product and then when that was finished you transferred the drawing to the production facilities where the Production Engineers decided how to produce it. The design process was seen as a linear process where all the tasks where seen as separate tasks that was dependent on the task that was performed in an earlier step. The design was also seen as a thing that when finished, it was “thrown over the wall” (Figure 1 top part) to some other task and it was not your problem any more. Today it is quite obvious that this was not a good thing to do since errors that was discovered in the design at later stages in the process was very costly since a lot of things had to be redone. In order to cope with these problems and try to avoid costly late design changes a Concurrent Engineering design process (Figure 1 lower part) have been used extensively. What’s so special about the design process is that multiple task are examined and designed concurrently, for example engineers from the design, analysis, manufacture and logistic tasks may work close together in order to design the best possible product that not only suit the customer needs but also the internal company needs and demands as well as to fit inside containers for excellent logistics. Using these kind of processes usually involves using computer aided tools and methods that may be quite complex and advanced however some of the methods are leaning more towards new ways of thinking and work methods that can be adopted with only small investments. One of the stated advantages with using concurrent methods is that you decrease the time it takes to bring the product to the market.

Below a couple of diagrams are shown showing the difference between traditional “over the wall” product development and using Concurrent Engineering methods. As well as generic product development processes according to common literature within the field.

Figure 1. Picture Adopted from Prasad, B., “Concurrent Engineering Fundamentals: Integrated Product and Process Organization”. Vol.1., pp.92, New Jersey, NJ: Prentice Hall, 1996.

Figure 2. A generic product development process. Adpoteb from Ulrich, K.,T., Eppinger, S., D., “Product Design and Development”

Figure 3. Steps of planning and design process, adopted from Pahl, G. and Beitz, K. “Engineering Design – A systematic approach”, 2nd Ed. Springer Verlag

Figure 4. The Mechanical Design Process adopted from Ullman, D., G., “The Mechanical Design Process”, 4th Ed. pp.82, 2010, McGraw-Hill

One of the key differences between the sequential traditional model and the Concurrent Engineering one is that Product Development is seen as an iterative approach where you can go back and forth and redo things until it is satisfactory before you build anything instead of just blindly going forward until you can’t go any further due to some unsolvable problem.

All these design processes can look quite daunting and complex but are actually more or less the same however presented from different perspectives. Below a few of the product development process tasks are explained. For more detail you can borrow some of the recommended literature from the library or search the web on the terms that have been mentioned.

Product Discovery, Product definition and Planning

In product planning and product discovery a lot of the background information and searching is done.

One can usually start off the design process with a need or a problem that needs to be solved. A need is more beneficial to have in the beginning since this need may exist at other places in the world and thus may result in both a more successful product and a bigger market. Usually you get an idea of something that needs to be solved but is your needs the same as the needs of the rest of the world? Well let’s say that you find a real need and that you have a hunch how you can satisfy that need. So how go about designing it? Well, will try not to use the ad-hoc methods that let you wildly rush through things without any thought of later stages in the product life-cycle. You will certainly end up with something but it may not inherit the properties that will suit the need both from a user perspective and from a business perspective. So in the next few section we’re going to go through a couple of steps that should be included in the design process.

First think about the product. Is it something that YOU think is excellent and solves a need (technology push) or is it a need that the market have asked for (market pull) or is it a new variant of an existing technology or product (product change). Depending on the situation you should adapt your development process and check the ‘real’ market situation, company needs and the economic outlook. You should allocate resources, material, personnel and money for the development project. THis is important so that you can afford to invest time in it. Create a product proposal and clarify the task. Also try to elaborate a requirements list for the product that reflect on the found needs.

The requirements list can be mentioned a bit more. Usually when you here the requirements list you think that you have to know everything about the product before you can start. If you do it’s good, but often you feel that some aspects are very clear but some other aspects are not known at the current state in time or are determined by other things surrounding the product or design itself. Don’t worry! You can always start with the ones that you know and state them as precise as possible. Try to use measurable metrics based on facts (i.e weight, speed, colour, mechanical properties etc.) since this will help you evaluate the design at later stages in the development project. You can always return to the requirements list and add or remove or change things as you fin out more about how the final solution. It can also be quite difficult to transfer the needs or need statements to measurable metrics, Sometimes it’s not possible. The important thing about trying to keep a connection between the needs and the requirements is that it gives you a broader understanding of why a requirement is set to a specific measure. This will affect your decisions later as some needs are seen as “more important” than others. This is important to think of during the entire design process to keep track of why things are designed the way it are. This is called Design Rationale.

Benchmarking, that is searching for products that are very similar to what you are going to design and examine those in great detail to position in the right market segment or see how your product features relates compares to the competition is important. Searching patent databases to avoid solutions that are patented or perhaps incorporate patents (you buy the patent or pay a license fee to use it) that may enhance it is always a good idea.

Trying to examine related technologies is also a good idea at this point. Perhaps there is some technology or product from a completely different field that may be used within your design or that you can be inspired from. Examining nature lies within this field and have resulted in for example the “Velcro” which was inspired from a plant.

Create a mission statement document that explains why, what and how you are going to realize this product. Make a project plan with specific tasks and a timeline that explains when things are supposed to be finished. GANTT-charts are quite useful for this.

Conceptual Design, Concept Development

When you think that you are finished with the planning phase, you have a clear understanding of what it is that you should accomplish, a plan how to realize it and some needs and requirements to act as your guide along the development, it is time to start creating some conceptual design.

A lot of the task performed in this stage is involved at abstracting the design in various formats. This is done to force you from fixation on a specific technology and conventional ideas. This means that although that your business e.g. has used hydraulics extensively in previous products it is not a necessity to use it again in all future products. It is a process of emphasising what is general and essential and ignoring what is particular or incidental.

Concept generation involves creating block diagrams of i.e. how the product layout will look like. Creating these charts enables you to analyse the product and it’s inherent components how they relate to each other. The focus on these charts should be of what kind of functions and behaviours the product should inhibit and not focus so much on technical solutions.

An important schematic is to draw the overall functional structures and break it down into more and more detailed sub-functions. Start at the System-level and break it down into sub-functions that build up your product (Top-Down). Or start off with the sub-functions that you think you need and then create the system-level design (Bottom-Up). A guideline is to think What and not how. This can be quite hard since we tend to think in technical solutions that we are aware of (how things can be solved) instead of what needs to occur (motions, cooling, heating etc). Try also to think in a timeline, what things have to occur in what order, or in some other way categorize them in a logical manner. Remember that this is an iterative process and that you may have to redo things a couple of times before you or the team are satisfied. Also remember that you can always go back and refine and change things.

After different systems and sub-systems have been identified it might be good to examine what kind of working principles that can be used to fulfil the functions of these sub-systems. Examples of working principals for the sub-function “Energy Storage” are electrical storage, mechanical storage (flywheel, moving mass), chemical storage, hydraulic solutions (bladder, piston, Potential energy).

Creative Methods

Creating concepts is a very creative process and there exist numerous methods that you can use to trigger yourself ti think in new patterns. Some of the methods used are:

• Brainstorming
• Methods 6-3-5
• Using analogies
• Using reference books and trade journals or whatever materials that you can be inspired from
• Experts – use experts from the desired field that can inspire or can come with input to your concepts. This may also involve the desired user group.

The concepts that you create may be sketches, physical prototypes of parts of the product or system or other types that explains how to solve a problem or suit a need from the requirements list.

When a number of concepts for the total system or for smaller more specific sub-functions have been created. You can try and examine if you can combine them in some way. Perhaps you can group things into a module that reduces the number of parts in the final product.

Concept Evaluation

After you have generated a few concepts it’s time to evaluate them. Now this can be done at various levels ranging from high fidelity (very precise and with low uncertainty) to low-fidelity (gut feel, best guess etc). However for the concepts to be compared they must be on the same level of abstraction, the alternatives as well as the criteria to be evaluate must be in the same language. You can’t compare apples with pears! Try to use measurable facts and figures.

Also in the beginning of the concept design phase there is a lot of uncertainty and as you start refining the concepts in greater detail you may discover that a property that you initially thought was good changes it’s behaviour during detail design. However there is always uncertainty in the process you just have to be aware that this is the case and you should have a plan on how to deal with it.

If you have used different types of modelling techniques (physical, proof of concept or virtual models) you can perhaps use them to analyse the functionality and the criteria you have set up.

Feasibility evaluations are a way to decide to go ahead or not with a technology or a concept. They are often used in the initial stages of concept evaluation. Common topics are:

“It is not feasible”, “It is conditional acceptance”, “It is worth considering”.

It should be mentioned that concepts that seem not feasible to continue the development should be looked at from other perspectives before discarded.

Other more systematic ways of evaluating concepts is via different types of Matrixes. One example of these matrixes is the decision Matrix (Pugh’s method). It is a method that let’s you compare alternative concepts in terms of different criteria. It’s a method of scoring each alternative relative to others in its ability to to meet the criteria. Comparison of the scores gives you the insight in what concept that is the best solution from a specific perspective. For more information regarding this search on the web for Pugh’s matrix

Another quite common method to evaluate products and concepts is called Quality Function Deployment (QFD). It is quite powerful if used correct but is very complex to understand and to go through. In short it is a combination of different matrixes that let’s you correlate different requirements and functions within the product and evaluate them in terms of different criteria.

Product Development/Detail Design/Embodiment Design

In this stage it means going from the concepts that you have created into even more detailed analysis of the parts or other components within the product or design. It aims to fulfil a given function with the layout, correct component shapes and materials.

This step may be a bit complex since it involves doing many things simultaneously, some steps have to be repeated at higher level of information and additions and alternations in one area may affect other components in another area.

In this phase you should consider Design for X (DfX) where X means analysing and adapting the product from different perspectives;

• Safety
• Laws and Regulations (CE-marking, ISO-9001 and ISO-14001 certificates etc.)
• Environmental impact
• Ergonomics
• Production
• Assembly
• Maintenance
• Operation
• Recycling
• Costs
• Logistics

It is important that you work in a systematic way in order not to forget something that may affect the end design negatively. Basically you look at your concepts that you still consider options for your solution and try to analyse them and adapt them according to the above topics. While you analyse your concepts you will stumble upon contradicting facts and problems, e.g. some manufacturing methods would reduce the cost of the component or product but requires a redesign of a sub-part, which will reduce the availability when assembling the product which will add to the total cost of the product. Any combination will suffice but a decision has to be made of which way to go.

However although all this analysis is done you still have to make decisions thus decision making is a large part of product development.

Decision making

During the design process we make decisions. We take these decisions based on fact, intuition, gut feeling or just plain foolishness. Well, we often try to avoid the foolish ones, but sometimes those are needed as well and may result in really successful products.

Decision based on fact are those that we “know” are in a certain way. For example when choosing between a number of different concepts that all solutions to our problem some are “better” or more “efficient” etc. in some measure. That is, you have criteria that you have set up that needs to be analysed in the concept and you choose the concept that has the highest rating of that measure. It is important that these measure are quantitative, that is, you can put a figure on it, without guessing the measure. Quantitative decision making is usually not done in the early phases of product development or the design process, often due to the fact that we don’t know enough about our design to state any important measure that let’s us compare things. However as the design and product progresses and becomes more and more detailed it is necessary and important to not base decisions on gut feel but on fact. This is much harder than it seems but very important.

I’ll try and make an example of the difference between gut-feel decision and qualitative ones. Say that you make three different concepts up that you intuitively see solves your problem. However which is the best or least bad? They all solve the problem? Well here comes the systematic part into consideration. In order for you to evaluate the different concepts you have to decide criteria’s that can be evaluated against each other. For example you can evaluate the concepts by the number of operations that’s needed in order to produce it, and then you can evaluate the manufacturing cost for that part as well since you know what operations and what machines that are needed. You can also evaluate it in terms of logistics and transportation. Which concept has the most parts that can be transported inside a standard shipping container, are they to big or inhibit a strange geometry that makes them hard to stack or package? These things add to the transportation cost. How can the different designs fit or mate to other parts in the product or assembly? Is there any mechanical properties that are needed for a certain design? Can we exchange material to another quality or type?

This kind of thinking is often called Design for X, where X can be Manufacturing, Transportation, Assembly, disassembly, Maintenance, exchangeability. Also to take into account is the number in which the product is going to be produced where small changes in the design can make real difference in the price for the end product. Also the number of units produced also affects the time that you can put into the design phase if you are going to be able to earn some money on it.

There exist numerous methods that can be used when taking decisions. Quality Function Deployment (QFD), Pugh-matrixes, TRIZ, Dot-voting and others. Often they both support both gut-feeling decisions as well as qualitative ones.

By analysing and using systematic methods when examining the part you can adjust and adapt the product according to your specifications. Using engineering methods and modern analysis tools as a complement to traditional manual tools and methods you can design, analyse and redesign your product in a quick way. Below some of the most common tools and methods are mentioned.

Computer Aided Design (CAD)

This is heavily related to creating 2D and 3D virtual models of your product. 2D drawings are rarely used anymore within larger companies, except in the manufacturing phase used as reference, 3D models are much more common. The reason for this is that 3D models make it easier to evaluate and design the shape of the parts and their interaction to other components, since you can twist and turn and zoom into the design. Another big advantage is that the 3D models can be re-used in different ways when analysing physical properties of parts and their interaction, as well as to adapt the manufacturing and production facilities completely within the computer. This way of thinking is called Master Modelling where you have only one design that you make small adjustment to when analysing. Some of the modern tools to analyse the product is mentioned below. A few of the modelling and analysis methods are;

• Finite Element Analysis
  • Often used for structural analysis of mechanical components but the method is so general and used as well in thermodynamic and other physics analysis. It is quite easy to grasp how it works and functions for easier but demands quite a lot of understanding to master.
• Multi-body Dynamics Analysis
  • This is basically dynamic simulation of mechanical systems such as cars, trains, engines, etc. Basically you create assemblies of parts (CAD-models) that you connect together using different joint such as revolute joints etc. you add material properties to the parts and then you run a simulation during a set time-frame. The software calculates forces, momentum and other interesting propoerties that may be hard to calculate by hand. Often the results from these kind of simulations can be used as input to Finite Element Analysis.
• Computational Fluid Dynamics
  • Often used to analyse different fluid and their interaction with components etc. Different modeling techniques exists.
• Thermodynamics
  • Programmes that lets you analyse thermal effects on components ro systems within your product. Different modelling techniques exist.
• Electric schematics
  • Electronic CAD-software lets you not only create the schematics for the electronics it also can help you analyse the electronic system and test it out virtually before any physical prototype is built. It can generate circuitboard design as well as 3D-modells of the components and circuitboard.
• Mechatronics
  • Lets you create a electro-mechanical system of both rigid parts as well as electric or other components. It enables you to design how the sensors and controls should behave and examine that the mechanical system works the way you want.
• Production cells
  • If you have your product design in 3D you may also incorporate its manufacturing and assembly in a production line. Where you can examine ergonomics (using manikins) and also design robot cells etc.

An umbrella term for these kind of tools and analysis methods is Computer Aided Engineering (CAE) or CAx. Some of these tools involve having engineering and mathematical skill in order to utilize them to it’s fullest however some of them are becoming more and more common and easy to use. Also it has become quite usual that you can download and use evaluation versions of the software. Often universities and higher education have student licenses so that you can use the software within specific courses. Also quite good software’s are available for free if you just want to learn basic skills. Also some of the bigger CAE-systems that are commercially available have many of these functionalities built within them or can be added with modules.

Keywords

Product Development

Design

Design Methods

Engineering

Engineering Design

Concurrent Engineering

Quality Function Deployment

Pugh Matrix

Design for X (DfX)

Computer Aided Design

Computer Aided Engineering

Finite Element Modeling

Thermodynamics

Computational Fluid Dynamics

Multi Body Dynamics Analysis

Mechatronics

Litterature

There exist great numbers of literature that explains this topic from several different perspectives and some of them are mentioned below:

[1]        Pahl, G. and Beitz, K., 1996, Engineering Design : A Systematic Approach, Springer, Berlin.

[2]        Prasad, B., 1996, Concurrent Engineering Fundamentals: Integrated Product and Process Organization, Prentice Hall, New Jersey.

[3]        Prasad, B., 1996, Concurrent Engineering Fundamentals: Integrated Product and Process Organization, Prentice Hall, New Jersey.

[4]        Pugh, S., 1990, Total Design : Integrated Methods for Successful Product Engineering, Addison-Wesley, Wokingham.

[5]        Cross, N., 2000, Engineering Design Methods : Strategies for Product Design, Wiley, Chichester.

[6]        Ullman, D.G., 2003, The Mechanical Design Process, McGraw-Hill Inc., New York.

[7]        Ulrich, K.T. and Eppinger, S.D., 1995, Product Design and Development, McGraw-Hill Inc., Boston, MA.

[8]        Andreasen, M.M. and Hein, L., 1987, Integrated Product Development, Springer-Verlag, UK.

[9]       Journal of Engineering  Design