IS as optimization problems
Posted: Wed Feb 12, 2025 10:23 am
An optimization problem considers one or more decision variables, whose values define the search space. This space is explored in order to find those values that offer the best evaluation of the objective function(s) that we want to minimize or maximize. These objective functions are often in conflict, that is, the optimization of one of them implies the degradation of the others. Below I will present three common problems in SI and describe the variables, the search space and the objective functions to be optimized.
Problem with the next version
Ideally, the next version of a product should consider all requirements from all customers argentina phone number data who have given feedback on the current version, in order to keep customers and, above all, to keep them satisfied. However, not all requirements can always be taken into account, often for cost reasons. In this case, the following question arises: what is the subset of requirements that balances software development cost and customer satisfaction? In this problem, the decision variable is a subset of requirements and we have to consider all possible subsets. For example, if the number of requirements is 50, the number of possible subsets of requirements, i.e. potential solutions, is 2 50 , which is approximately 10 15 . Considering all these subsets to find the one that offers the best assessment of the objectives is practically impossible.
Problem with the next version
Ideally, the next version of a product should consider all requirements from all customers argentina phone number data who have given feedback on the current version, in order to keep customers and, above all, to keep them satisfied. However, not all requirements can always be taken into account, often for cost reasons. In this case, the following question arises: what is the subset of requirements that balances software development cost and customer satisfaction? In this problem, the decision variable is a subset of requirements and we have to consider all possible subsets. For example, if the number of requirements is 50, the number of possible subsets of requirements, i.e. potential solutions, is 2 50 , which is approximately 10 15 . Considering all these subsets to find the one that offers the best assessment of the objectives is practically impossible.