An Enhanced R-NSGA-II For Multiple Brands Advertising Campaign Allocation Problem

Abstract

This paper deals with the Campaign Allocation Problem of commercial Ads in TV breaks that we formalize as a multi-stakeholders multiobjective problem with highly competing objectives for different brands and numerous constraints. The problem is NP-hard with a high dimensional objective space and scalability issues in terms of the number of breaks. Moreover, the expected solution should be able to focus on a sub-part of the Pareto front according to decision maker's (DM) knowledge. To tackle these challenges, we propose to use R-NSGA-II, a Many-Objective Evolutionary Algorithm (MaOEA), combined with a novel gene encoding/decoding process. Experiments show that this approach obtains better results than usual MaOEA (NSGA-II, NSGA-III) according to industrial performance criteria, scales to large instances, and incorporates decision maker's preferences during the optimization process.

A new reference-based algorithm based on non-euclidean geometry for multi-stakeholder media planning

“…Results in this section have confirmed our findings that the use of R-NSGA-II as proposed by Benali et al in [1] does not seem to be the most appropriate choice for our industrial problem presented in Section 5. As R-NSGA-III is better than R-NSGA-II in terms of diversity and convergence, it will be used for comparison in the industrial case study experiments.…”

“…Inspired by the work of Benali et al [1], our idea is to use an EA to find Pareto-optimal solutions to the Campaign Allocation Problem. We describe here the main contributions in Multi-Objective Evolutionary Algorithms (MOEAs) and Many-Objective Evolutionary Algorithms (MaOEAs) before emphasizing on the importance of reference points to guide the convergence in high dimensional objective space as the case of our problem of Multi-Stakeholder Media Planning.…”

“…Unlike the previous contributions, our work in [1], considers multi-stakeholders configuration, i.e., advertisers, and TV networks, each with their own objectives. They also consider the size of the problem instance, all breaks from a 3 to 6 months period, that may not be appropriate for an exact resolution based on solvers.…”

“…According to their problem which is a sub-part of ours, Benali et al [1] use for instance R-NSGA-II [13] to target their objectives. Despite the fact that their proposal obtains interesting results, they tackle the problem only partially.…”