Designing a Data-Driven Decision Model for Optimising Hotel Operations: An Integrated Study of Demand Forecasting, Workforce Allocation, Queueing Systems and Resource Utilisation

Ruveer Sarwal, Raghu Raja Mehra

Abstract: Hotels take four important decisions every day: how many guests to expect, how many staff to put on duty, how many service counters to open, and how much of the building to keep running. In most hotels these four decisions are taken separately. This paper builds a single decision model that links them together. A forecasting model predicts tomorrow's room demand, the forecast decides how many staff each department needs, a queueing model checks whether the planned counters will keep waiting time below a target, and a resource index tracks how well rooms, staff, energy and water are being used. Because hotels rarely share day-by-day operating data, the model is tested on a clearly labelled simulated dataset of 365 days for a 180-room hotel; the data are artificial and are not taken from any real hotel. On a 60-day test period, a random forest gave the most accurate forecast (average error 6.34 rooms, 5.59%), just ahead of Holt-Winters smoothing. Replacing a fixed roster with the forecast-based plan raised staff utilisation from 81.7% to 91.8%, cut the average check-in wait from 11.34 minutes to 2.00 minutes, reduced daily staffing cost by 13.0% and improved the resource index from 72.6 to 79.5. Most of the improvement comes from moving staff to the busy hours rather than from hiring fewer people. The results are illustrative and would need testing on real hotel data before they can be generalised.

Keywords: hotel operations; demand forecasting; staff scheduling; queueing theory; resource utilisation; decision model; simulation

How to Cite?: Ruveer Sarwal, Raghu Raja Mehra, "Designing a Data-Driven Decision Model for Optimising Hotel Operations: An Integrated Study of Demand Forecasting, Workforce Allocation, Queueing Systems and Resource Utilisation", Volume 15 Issue 8, August 2026, International Journal of Science and Research (IJSR), Pages: 2134-2143, https://www.ijsr.net/getabstract.php?paperid=SR26825124152, DOI: https://dx.doi.org/10.21275/SR26825124152

Download Citation: APA | MLA | BibTeX | EndNote | RefMan