Speaker
Description
The construction of sports complexes often aims to host large sporting events and foster urban area development (Baade and Dye, 1988). During the post-event period, these complexes must maintain economic gains by providing commercial facilities with sports and living services. However, the poor performance of facilities has led to a waste of urban resources. The facilities' profitability became a key indicator when assessing the sustainable utilization of sports complexes. With rising public health awareness and demand for sports activities, there is still a need for new complexes. Thus, investigating the factors influencing the profitability of commercial facilities is vital for their effective planning and operation.
Previous studies have examined various factors determining commercial facility profitability, such as location, accessibility, commercial layout, and operational hours. Various models have been applied to examine how these factors affect profitability, such as AHP (Erturan-Ogut and Kula, 2023) and linear models (Tan et al., 2022). However, most existing models require the construction of evaluation weights based on subjective evaluations or ignore nonlinear relationships, which may lead to biased or invalid results. Recent advanced analysis techniques, particularly machine learning methods, enable more accurate estimation. In previous research, machine learning models were applied to predict the sales potential of hospitality services based on several factors of sales performance (Han et al., 2022).
In response, the study proposes a method based on a machine learning model to predict the profit potential of commercial facilities within sports complexes. The study selected several commercial facilities within 26 sports complexes in Shanghai for empirical research. The price per customer of these facilities was selected as the dependent variable to represent the profit potential, with other influencing factors serving as the independent variables. The machine learning model, Random Forest, was applied to build the model. The study's findings indicated that the location and operational hours of commercial facilities within the complexes significantly impacted their profit potential. The study further employed a cross-validation approach to refine the model and evaluate its robustness. The findings demonstrated the model's capacity to accurately estimate prospective planning facilities' profit potential.
The empirical results demonstrate the efficacy of the proposed method in assisting sports complex operators and urban planners in evaluating the potential profitability of commercial facilities. This assessment enables informed planning and operational decisions, which promotes the sustainable utilization of sports complexes in the post-event phase.
References
Baade, R.A. and Dye, R.F. (1988) ‘Sports Stadiums and Area Development: A Critical Review’, Economic Development Quarterly, 2(3), pp. 265–275. Available at: https://doi.org/10.1177/089124248800200306.
Erturan-Ogut, E.E. and Kula, U. (2023) ‘Selecting the right location for sports facilities using analytical hierarchy process’, Journal of Facilities Management, 21(5), pp. 733–750. Available at: https://doi.org/10.1108/JFM-09-2021-0103.
Han, S. et al. (2022) ‘Search well and be wise: A machine learning approach to search for a profitable location’, Journal of Business Research, 144, pp. 416–427. Available at: https://doi.org/10.1016/j.jbusres.2022.01.049.
Tan, X. et al. (2022) ‘Performance Optimization Method of Community Sports Facilities Configuration Based on Linear Planning Model’, Complexity. Edited by H. Garg, 2022(1), p. 4489802. Available at: https://doi.org/10.1155/2022/4489802.
| Keywords | machine learning; commercial facilities; sports complex |
|---|---|
| Best Congress Paper Award | No |
