Year: 2026
Type: dissertation
Abstract: Retail promotion planning is an operationally challenging function. Despite the availability of dedicated enterprise software, planners still depend on manual workflows and, as a result, generate inefficiencies at scale. This thesis examines how Generative AI can be used to improve promotion planning and decision-making in enterprise software with a Nordic retail planning and supply chain software company as a case study. A qualitative case study was conducted using semi-structured interviews with nine domain experts from product, consulting, and technical roles, including one retail end-user. The analysis revealed seventeen challenges, which were distributed across five categories: process inefficiency, user experience, data accessibility, decision support, and integration. The most common issue cited was bulk promotion creation (eight out of nine participants). Creating a single promotion requires approximately twelve interface steps and two to three minutes, which is a time commitment that is prohibitive when scaled up for hundreds or thousands of promotions per planning cycle. Each challenge was mapped using a priority-weighted approach to architectural solutions. Phase 1 focuses on the facilitation of bulk promotion creation, guideline-based editing and provision of evaluation summaries with actionable recommendations. All participants reported prior professional experience with Generative AI tools, 67% had no major concerns about adoption, and text-based interaction modalities were preferred. Privacy, data governance and trust in AI outputs became key design requirements. The thesis makes a contribution to a validated challenge taxonomy, a systematic challenge-to-architecture mapping, and an enterprise GenAI assistant architecture representing a design artefact grounded in the empirical findings.
Source: Aaltodoc (Aalto University)
Author: Hussain, Asad
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7162225133