EVOLUTION OF METHODOLOGICAL APPROACHES TO FORECASTING BUSINESS PROCESSES IN THE SERVICE SECTOR: A SYSTEMATIC REVIEW FROM THE PERSPECTIVE OF AN OPERATIONAL MANAGER
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The relevance of this study is determined by the urgent need to bridge the gap between the theoretical arsenal of forecasting methods and the real possibilities of their operational implementation in the highly dynamic service sector. In conditions where an error in demand forecasting is directly converted into financial losses and a decrease in the quality of service, the choice of an adequate method becomes a critically important strategic decision for a manager. The aim of this work is to conduct a comprehensive systematic review and a multifaceted comparative analysis of existing methodological approaches to forecasting business processes in the service sector, assessing them through the prism of operational applicability, practical barriers and ultimate value for the decision-maker. The methodological basis is a systematic review of scientific literature over the past five years, supplemented by a comparative analysis based on the author's system of criteria, developed taking into account the realities of operational management. As a result of the study, an original taxonomy of methods has been developed, and a system of fundamental trade-offs between their key attributes has been identified and analyzed: stability and flexibility, scalability and localization, complexity and manageability. It has been proven that in the context of services, forecasting efficiency is largely determined by the method's resistance to operational risks (human, technical, contextual), rather than its pure mathematical accuracy. The key conclusion is the justification of the hybrid approach paradigm, which involves not the choice of a single "best" method, but the targeted integration and orchestration of three classes of methods — time series, expert assessments and machine learning — within a three-level architecture of a forecasting system. Practical recommendations for building such systems have been developed and detailed, supported by the results of an implementation case study in a fast-food restaurant chain. The scientific novelty of the study lies in the operational and managerial focus on the forecasting problem, which makes it possible to translate theoretical discussions about the accuracy of algorithms into the plane of practical decision-making, taking into account the constraints of the real business environment.
Fields of Study & Subjects
Forecasting Techniques and Applications
Big Data and Business Intelligence
Innovation Diffusion and Forecasting
Publication details
- Journal / source
- EKONOMIKA I UPRAVLENIE PROBLEMY RESHENIYA
- Volume
- 2/7
- Issue
- 167
- Pages
- 187–200
- Publication date
- 1 January 2026
- Language
- EN