Product simulation and testing

From WikID

What Is Product Simulation and Testing?

Schematic representation of product simulation and testing

Product simulations and testing aim at gaining an understanding whether the product functions the way it is intended to do. New product ideas and concepts are created through finding and describing the functions and the use of those functions. The functions are materialised/embodied with technical solutions principles. Designers try to find the best technical solution principle that can make a particular function, or set of functions, work. Several (existing) technical solution principles are possible for a (set of) function(s), and sometimes a new technical solution principle must be found. In the creative phase of the design process, it is your job to find the most appropriate technical solution principle for the desired function(s). Simulation plays an important role: in order to judge the solution principles found, you have to determine the ‘quality’ of your design and gain insight into the functioning of your design through simulation.

In order to perform a simulation, you first need to construct a model of the desired function and technical solution principle. A model is a simplified representation of a real-world phenomenon, which is not reality itself, but can be used as a way of describing, explaining and making predictions. Within the design process, many different types of models can be used for simulation and testing purposes: dummies, mock-ups, prototypes, but also drawings and diagrams. Using the models, you can test your assumptions; modelling allows for experimenting and testing whether the solution principles behave as intended. The process described above shows great similarities with the ‘scientific method’, which is typically used in scientific research (see also Roozenburg and Eekels, 1998). In this context, design can be seen as a process of making predictions. First, designers hypothesise about how a certain technical solution principle fulfils one or several predetermined functions. Next, they construct models to make predictions about this process and through simulation with the model, they investigate whether the predictions sustain the hypothesis. Experiments are then needed to validate the model and check whether the accuracy of the predictions is sufficient. In other words, through experimentation, designers determine whether the developed model proves that the principle behind the product or function is indeed as they had hypothesised. Modelling, simulating and validation through experimentation are important aspects of the design process.

Models

Models can be classified in various ways. Here, we distinguish between material and symbolic models. Material models are various sorts of prototypes, such as sketch models, detailed models, dummies, mock-ups and final models. Symbolic models are diagrams and mathematical models. Another classification of models is according to the type of simulation: (1) simulation with structure models, (2) simulation with iconic models, and (3) simulation with mathematical models.

Example of a structure model
Example of an iconic model
Example of a mathematical model
  1. Examples of structure models are flow diagrams, circuit diagrams and function block diagrams. Sketches and dummies are also included in this group. Structure models are qualitative and are used to assess the qualitative structure of a product or a process. They give a quick first impression of the appearance and functioning of the product. Structure models are often the first step to more advanced models.
  2. Examples of iconic models are pictures, drawings, dummies, mock-ups, and prototypes. Iconic models have a similar geometry to their design: simulation with iconic models is more realistic, concrete and quantitative. Three-dimensional models form an important group: dummies, mock-ups, sketch models, detailed models and prototypes. Functional prototypes enable designers to test the functionality and usability of the design with a high degree of realism.
  3. Examples of mathematical models are mathematical formulae, such as Newton’s law, to determine the physical characteristics of the product. Mathematical models can be used to evaluate the physicochemical parameters of the design in question. These models help you to quantify and determine the parameters of the components and the dimensions of the product. They give an objective view on the problem in hand and the results are fully quantitative .

Simulation

Example of simulation with a mathematical model (graduation project Marco Koekoek, TU Delft, 2007)

By means of models, described in the previous paragraph, you can perform different simulations, depending on the information required. The questions that you try to answer could be as follows:

  1. What constitutes the function that the product must fulfil?
  2. Does the product perform as intended; will it fulfil its functions?
  3. Can the product be manufactured in the planned quantity, and at an acceptable quality and price?

The following list provides some examples of simulations for ‘answering’ specific questions. These particular simulations have become well-known thanks to their extensive use in design practice:

1 Failure models and effects analysis and fault tree analysis

Failure models and effects analysis (FMEA) and fault tree analysis are two qualitative methods for analysing the reliability of a new product. Applied early in the design process, they can help you to find the possible causes and effects of failure. Through FMEA an answer to two questions is sought: (a) in which manner can the part fail, and (b) what happens if the part fails? The result of the analysis is a list of critical points and an indication of what should be done to reduce the chance of failure. In a fault tree analysis (a structure model) you look for the causes of a presumed failure mode of the product. The advantage of fault tree analysis is that it indicates how the reliability of a complex product depends on the functioning of the separate parts.

2 Experiments with prototypes (material models)

In early phases of the design process some insight needs to be gained, in order to be able to abstract a function or product into a mathematical model. What factors are relevant is often not known in advance and will become apparent in practice. Given this experience, further investigation can be performed to find out which parameters have an important influence. Also, at the final stages, proof of principle of critical parts is often tested in a trial set-up, using detailed or final prototypes. They play an important role in the simulation of the manufacturing process to discover lacking features. Then the dimensions of the product have to be completely defined.

3 Finite element method (FEM)

Science provides a variety of mathematical models to describe physical phenomena. FEM is an example where the mathematical model becomes so complex that the simulation can no longer be done by hand. The principle of FEM is that an object or system is divided into small cells. The interaction between two aligned cells is modelled through the laws of nature. Depending on the level of detail, the number of cells is large and calculations have to be automated. Several computer programs are available that can apply FEM on a geometry. However, the models or the form of the cells used in these programs are often hidden. Therefore, a critical view on the outcomes is important and should ideally be checked through (simple) manual calculations.

4 Scaling up to mass production

At the end of the design process, only one product is designed. When mass-produced, this product needs to be modified. By means of prototypes and trial runs, the product can be prepared for mass production. Usually a prototype is followed by a trial run of a batch, the null series, to see if no problems occur during production on a large scale.

5 Logistics and quality analysis

During the manufacture of a product, materials, parts, and subassemblies ‘flow’ from one workstation to another. These material flows can be visualised and analysed using network-like graphic models (analogue models) such as in ‘routing analysis’, the `Sankey diagram’, and ‘failure rate analysis’.

6 Design for Assembly (DfA)

A widely known and applied analytical tool is Design for Assembly. The assembly process of products is simulated by means of a mathematical model in the form of a system of tables that connects form features of parts to the estimated assembly time.

7 Value analysis

Value analysis is the analysis of the functions and subfunctions of a product, and the comparison of the value of those functions with regard to their costs. For that purpose the value of a function is equated in principle to the price of the cheapest ‘carrier’ of that function available in the market. By systematically setting values and costs off against each other, you can see which parts of an existing product, or new product design, are likely candidates for improvement. Unfortunately, value analysis is often wrongly associated with cost reduction only and not with quality improvement at equal costs.

8 Ergonomic Simulation

Example of an ergonomic simulation

Designers want to know what kind of user behaviour their design provokes, so that they can improve their design, if necessary. As you never have the whole population for which the product is intended at their disposal, a model of the design is tested on a ‘man model’ (mannequin). A mannequin is a representation or imitation of ergonomically relevant features of a certain population. The most important man models used in anthropometric ergonomics are tables and layout drawings (of work spaces), two-dimensional manikins, computer models of human beings, and test subjects.

9 Business-economic simulation

The attractiveness of the business potential of a design is another type of simulation that plays an important role in a design process. Cost price calculations are often made in a design process. For cost price calculations, most of the design must be known: what type of components, materials and production techniques. With cost price calculations it is possible to make profitability calculations. In profitability calculations the general profitability of a design project is calculated, on the basis of which calculation a go/no-go decision can be made regarding the continuity of the project. Forecasting methods help in making a prediction about the number of users that will buy the new product.

10 Social and Ethical Simulation

You could formulate social and ethical criteria in the design specification, and take these into account using the various decision methods presented in section 2.3. and applying social and ethical simulation. These simulations are not performed by means of mathematical formulae or experimental methods, but by means of conscientious thinking, logical reasoning, and common sense. For that purpose check lists can be useful, as can be found in Roozenburg and Eekels (1992).

11 Simulation of Environmental Effects

Product design always leads to unintended environmental side-effects in the production, distribution and use of materials. They are caused by the withdrawal of raw materials and energy, and emissions into air, water and the soil. You as a designer have an important role in decreasing the impact on the environment and creating sustainable products. In the design process, therefore, a designer should be interested in the impact of his/her design on the environment. In order to obtain a clear understanding of the environmental impact, a you could do an environmental effect simulation, for example by using a MET matrix. Another point to be taken into consideration is to deploy various EcoDesign Strategies, based on an analysis of the Product’s Life Cycle (also see EcoDesign checklist)

DDG-2-cart-product-simulation.png

When Can You Use Product Simulation and Testing?

Product Simulation and Testing take place throughout the design process, with increasing levels of concreteness, of detail and of accuracy of the models used. However, some types of simulation are applicable in the beginning of the design process, others near the end of it.

How to Use Product Simulation and Testing?

Howto-product-simulation.png

Starting Point

The starting point of simulation is an aspect, either a functional aspect or a material one, of the design that needs testing to verify its underlying assumptions on functionality, construction and materialisation. In other words, a feature of the design needs testing in order to prove its workings.

Expected Outcome

The outcome of simulation is a confirmation whether a particular aspect or feature of the design works or functions as intended.

Possible Procedure

Note: the following procedure is not necessary for all types of simulations described above.

  1. Describe the goal of the product simulation. Analyse the existing situation, and determine the various scenarios of use.
  2. Determine the type of model you will be using. Make the model; abstract the product idea into the symbolic language of the model. Build a prototype, if necessary. Select or construct the appropriate mathematical models.
  3. Carry out the simulation or test. Set up a plan for the test. Record the test and the results of the test.
  4. Interpret the results.
  5. Evaluate the results, and reflect the results upon the goals stated earlier. Also, reflect the result upon the initial product idea.

References and Further Reading

  • Roozenburg, N.F.M. and Eekels, J. (1995) Product Design: Fundamentals and Methods, Utrecht: Lemma.
  • Roozenburg, N. and Eekels, J. (1998, 2nd ed.) Product Ontwerpen: Structuur en Methoden, Utrecht: Lemma.
  • Christiaans, H. et al. (2004) Methodologie van het technisch-wetenschappelijk onderzoek, Utrecht: Lemma.
  • Otto, K. and Wood, K. (2001) Product Design: Techniques in Reverse Engineering and New Product Development, Upper Sadle River: Prentice-Hall.


Example of a simulation (graduation project Willemijn Verduijn, TU Delft / TNO, 2007


{{#css:/skins/DDG/DDG2.css}}

Personal tools
Aspects & Domains