What the paper studied
Udaipur, Rajasthan's "City of Lakes", has more than 400 hotels, from boutique heritage properties to international five-star brands, and its tourist season peaks in winter (October to March). Hotels there struggle to recruit and retain qualified staff, and labor market information for India's service sector is limited.
The authors surveyed 120 HR managers, department heads, general managers and senior supervisors at registered Udaipur hotels, selected by purposive sampling, over six weeks in October and November 2024. The 40-question instrument covered qualifications, training, recruitment methods, AI in recruitment, supply and demand issues, organizational culture and expectations for AI. Responses were analyzed with descriptive statistics and cross-tabulations, and the results informed a proposed AI forecasting framework.
Key findings
- No consensus on the balance. 36.7% saw manpower supply and demand as balanced, 33.3% saw supply exceeding demand and 30.0% saw demand exceeding supply, pointing to a market fragmented by property type, department and season.
- Shortages are uneven. 26.7% of respondents said their hotel never faces shortages, but 26.7% face them monthly and 23.3% weekly.
- Several problems at once. The top challenges were high employee turnover (27.5%), high recruitment cost (24.2%), shortage of skilled labor (24.2%) and seasonal demand swings (24.2%).
- Hiring is still traditional. Referrals (30.0%) and internal promotion (29.2%) are the most used recruitment methods, though online job portals were most often named the best method (34.2%).
- AI adoption is on a knife's edge. 49.2% of respondents use AI in candidate assessment, and the same share say AI helps with manpower forecasting. Views on accuracy are split: 22.5% rate AI forecasts highly accurate and 20.8% not accurate, and the average AI job-match rating is 3.03 out of 5. 34.2% expect AI use in HR to grow over the next five years, while the same share are unsure.
- No single fix. Industry-academia collaboration (27.5%), AI-driven forecasting (25.8%), better pay (25.0%) and flexible training (21.7%) were chosen almost equally as the most effective solution.
- Patchy data foundations. Only 51.7% of respondents reported structured job descriptions and KPIs, and 53.3% of hotels offer special training.
The proposed framework
- Labor market information hub combining state employment records, job portal vacancy and applicant data, tourist arrivals, hotel occupancy and revenue data, and hospitality graduate numbers.
- Demand forecasting with hybrid LSTM-GRU neural networks predicting department-level staffing needs six to twelve months ahead from occupancy, room rates, tourist numbers, GDP, inflation and events.
- Skill-gap analysis applying NLP and topic modeling to job descriptions and comparing them with local hospitality curricula and placement results, producing a skill gap index by department and education level.
- Optimization engine using mixed-integer linear programming to produce lowest-cost staffing plans that comply with labor laws, with short-term (0–3 months), medium-term (3–12 months) and long-term (1–5 years) recommendations delivered through a cloud dashboard.
Why it matters for hospitality
Seasonal destinations everywhere share the pattern this study describes: staffing shortfalls are noticed when they happen, not months before. The authors argue the answer is not only better demand forecasting but a supply-side pipeline that tracks graduates, skills and placements, so recruitment and training can start ahead of the season. Their point that forecast quality depends on basic HR data, such as structured job descriptions and KPIs, applies to any hotel considering AI workforce tools.
Practical takeaways
- Keep structured, regularly updated job descriptions and KPIs; they are the data foundation for any AI workforce forecasting.
- Pair demand forecasts with supply-side tracking of training output and skills so hiring can begin before shortfalls appear.
- Work with hospitality schools and use skill-gap data to shape curricula.
- Combine AI forecasting with non-technical levers such as pay and flexible training, since respondents saw no single solution.
At the policy level, the authors call for standardized manpower reporting by registered hotels and a labor market information hub run jointly by the local chamber of commerce (UCCI) and the Rajasthan Labour Department.
Limitations
The study covers a single city and relies on purposive sampling, a cross-sectional design and descriptive statistics, so the results describe managers' perceptions rather than causal effects. The AI framework is proposed but not yet built or tested; the authors suggest piloting it across 20 to 30 Udaipur hotels and tracking establishments over three to five years.