An Evolutionary Algorithm for the Rank Pricing Problem

This paper develops an evolutionary algorithm to solve the Rank Pricing Problem. In this problem, a company establishes prices of a set of products offered to a set of customers in order to maximise its revenue. The proposed algorithm exploits the property that there exists an optimal solution to the problem where prices take values over the set of different customer budgets. As prices can be discretised, pricing decisions become a combinatorial optimisation problem. The algorithm has three distinctive features: the definition of the chromosomes (based on prices), the generation of the initial population, and the application of a local search procedure aiming to improve the current feasible solutions. The computational experience carried out confirms the relevance of the algorithm, especially in terms of the computing time invested.