Random generation of potential design concepts

Finally, we are ready for the most exciting part of morphological analysis the generation of potential design concepts. These concepts may become the key to our fame and fortune. 

As we remember, we are looking for sequences of symbolic attributes and combinations of their values. Each attribute from the table must be used but only once in a given run. Also, for each attribute, only its single value must be randomly selected. For example, if a table with four rows is considered, a sequence of four symbolic attributes and their values may look like this  

We can produce random combinations manually or using various computer tools. Both paths are good using a computer tool saves time and nearly guarantees random generation, but it is also less fun. 

In the first case, we need to assume that each row is considered independently from the remaining rows and that the generation must be repeated for all rows. Let us consider the row i, which has m cells, each with a different value of the symbolic attribute A , which resides in this row. 

Our first step is to prepare small balls with numbers from 0 to k, where k is the maximum number of values in a row for the entire table. For our row number i under consideration, we will use only the first m balls, numbered from 1 to m, and we will put them in an open vase. Obviously, when we have an appropriately trained animal, a parrot or a monkey, we will ask our animal to draw a ball. Next, we will read the number on it and this number will become the number of a value to be used in our combination. For example, if the number is 4, that means that we will use the value Aj4 or the fourth value in the row i. If it happens that we do not have any trained animals, we should be able to draw a ball ourselves (but with covered eyes) and record the drawn number as in the previous case. 

---

The selected design concept was presented earlier in Section 6.5.4.1, Random Generation of Potential Design Concepts. ... 

---