ABSTRACT
The construction of a fuzzy B-spline can be done with a resolution of a linear programming problem. In fact, the pseudo code 1 shows a single statement to perform the resolution of the linear programming problem in Equation (2.17). It must be noticed that this problem can be hard to solve numerically because it could involve thousands of variables and constraints, but this is not unusual in linear programming. However, many traps could fail the computation. Basically correlations among geographical data (which are very frequent) can lead to linear combination and could block the start of many linear routines. An easy solution to these problems can be a simple exchange of the rows into the matrix of the linear problem. This easy heuristic can help the pivoting strategies.
