Product Design and Process Development
5.3 STEPS IN PRODUCT DESIGN AND PROCESS DEVELOPMENT
Carrying out the design in the five successive steps listed in Figure 5.1 goes some way towards eliminating the mistakes of choosing the wrong design and also making the product on a large scale when very little is known of the processing system.
5.3.1 'Getting the feel'
This is a continuation of the development of the product concept and the product design specifications. The processing methods and conditions outlined in the product design specifications are used to make the early product prototypes, and the technical testing methods are examined for reliability and accuracy in testing both the technical product characteristics and also their relationships to the consumer product characteristics.
There is a question of consumer involvement at this stage; some people advocate this strongly because it means that there is control over the design; others say that it is faster and just as accurate to use the knowledge of the designers. The choice of no consumer testing depends on the level of consumer knowledge held by the designer.
The basic costing used in the company is also identified so that a simple method of determining costs can be used in the next stages of the product design.
The target market was identified in the product concept stage and the consumers are selected to represent this target market(s).
Screening reduces the wide range of raw material and processing variables to the input variables affecting important product qualities. This hastens the design.
Initially the variables can be reduced using the previous knowledge of the designer and also published or company information easily available. There can still be a number of floating variables and these are studied in controlled experimentation, not 'ad hoc' try-and-see experimentation. Many experimental designs are available to screen the variables but the most common are partial factorial designs, or Plackett and Burman designs.
In a Plackett and Burman design, it is possible to screen N-1 variables with N experiments. The screening experiments identify the important variables and their magnitude levels that affect the product qualities, but they are not statistically accurate and cannot quantify the relationships between the input variables and the product qualities.
Some food designers have the consumers test many samples in these designs, sometimes for acceptability, but more usefully in product profile tests. Other designers use trained sensory panels.
At this stage, the raw materials are being selected, and the quality, availability and costs of those raw materials are studied. There is likely a basic total cost range for the raw materials, but it is important not to select individual materials only on cost at this stage. Higher qualities of raw materials may give a unique property to the product, and also the more expensive materials may not need to be used in the same quantities as the cheaper. Sometimes there are restrictions in the company on the raw materials that are to be used; the buying department can often give some indications without restricting the design.
5.3.3 Ball-park studies
In ball-park studies, the aim is to set the limits of the raw materials and the processing variables which give acceptable product qualities as judged by the consumer. By this stage, the variables are reduced in number and their outside limits are set. They are examined in factorial designs, and for raw materials in mixture designs.
In factorial designs each input variable is considered at high and low levels, and the combinations of these high and low levels for all input variables are tested. In a full design all possible combinations are run, therefore for three variables the total number is 23 = 8 experiments.
In food formulations, mixture designs are often used because it is impossible to vary one ingredient while holding all the others constant; in mixture designs, the sum of all the ingredients in the formulation must add to 100%. The product designer must always be aware that when they change the content of one ingredient, the proportion of the other ingredients changes, for example reducing the fat content will increase the proportion of other ingredients: carbohydrate, protein or water. With factorial designs and mixture designs, the effects of the various input variables, alone and together, on the product qualities are analysed, and mathematical relationships developed between the input variables and the product qualities. To set up the experimentation and to analyse the results, there is computer software readily available for food product development.
Both technical testing and consumer testing of these product prototypes are carried out. The consumers are testing for acceptability and the technical tests are examining the chemical, microbiological, physical and sometimes the sensory properties of the products. Accuracy and reliability are important considerations in this testing, both for studying the effects of the input variables on the product qualities and for developing the quality assurance programme. The total processing costs of these product prototypes are compared to identify the effects of the input variables on the costs, and to check that the costs are within the target cost range.
Here the aim is to optimise the overall product quality by determining the levels of the input variables which will give the best possible product quality. The problem is that often when optimising one product quality, another product quality is less than optimum. So it is a case of setting the relative importance of product qualities, and for the most important product qualities studying the formulation and processing variables to find the optimum. But the limits that are acceptable across all the product qualities need to be known so that during the optimising experiments none of the other product qualities become unacceptable.
For raw material formulations, linear programming can be used to optimise a number of product qualities and costs with the amounts of raw materials in the formulation held between upper and lower levels.
Scale-up (or ramp-up) of both the production and the marketing is the last stage of the product design and process development. The production scale-up is the in-plant test to verify that the product can be made at the quality and quantity required, and the marketing scale-up is a large consumer test to verify that the target consumers will buy the product and what marketing strategy will encourage this buying.
The aim of the processing scale-up is to determine the optimum production process for product quality, product yield, process control and costs. If the previous design research has combined the product and the process, this can be achieved without too many problems. But if the process has been ignored, then there can be disastrous problems. For example, if some of the intermediate materials have never been pumped during the design experimentation, then they could break down during scale-up.
The scale-up can be either on a pilot plant or short production runs on the main plant. If it is a new process, or there is to be quite extensive experimentation, then the scale-up is conducted on a pilot or small-scale plant. If the process is only an adaptation of the present production, then the scale-up is conducted on the main production plant. The decisions on the type of scale-up are often much influenced by cost; the production trial can cost a great deal if the product cannot be sold and this restricts the use of the production plant until the final stage. But if there is no investment money to build a pilot plant then the production run may be the only scale-up available.
The question can often be asked as to when the scale-up from the laboratory bench to the small plant to the production line should be carried out. A great deal of time can be spent perfecting a product in the laboratory, only to find that it is impossible to duplicate this in the plant. If the product is rushed from the laboratory to the production line, then there can be a great deal of raw material and product discarded at a substantial cost.
Knowledge of the interrelationship of the processing variables and the product qualities can reduce these failures. EVOP (evolutionary operations) are used in optimising the process variables, especially if using the production line in scale-up. EVOP is a way of plant operation that tests small changes in the process variables in a simple factorial design. It continuously changes the process variables until optimum product qualities are reached, but only slowly so that the product can be used for large scale testing or even sold.
The marketing scale-up aims to define the market, describe the market strategy to reach this market and predict the possible sales revenues for the product. Possible market channels are studied and the market channel suitable for reaching the target consumers and for the company is chosen. The price range related to the production costs, competitors' pricing and company policy is tested with consumers to see how it affects their buying intentions. Also the final product concept (the product proposition) is built up from the final prototype product, the packaging design and consumer studies. The definitions of the product, price and market channel are used not only to develop the aims and methods for the promotion of the product but are also the basis for planning the marketing mix during product commercialisation.
The final prototype product from the production scale-up and the various parts of the marketing strategy are tested in a large-scale consumer test where the consumers test the product in their usual environment and are interviewed about the marketing strategy.
In Case Study 5 are some comments on food design for the future from an Italian design journal, to start you thinking about the development of food design. Certainly today there is a need to determine the direction of food design both for food ingredients and consumer products, and also for the fresh products which are being designed on the land and in the water.
Back to the top