Workforce Forecasting with Machine Learning for Healthcare Management


Koç D. T., Eren E.

Journal of Health Management, 2025 (ESCI, Scopus) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Basım Tarihi: 2025
  • Doi Numarası: 10.1177/09720634251396339
  • Dergi Adı: Journal of Health Management
  • Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus, CINAHL
  • Anahtar Kelimeler: health economics, Healthcare management, healthcare policy and planning, healthcare supply and demand, machine learning, time series forecasting
  • Yıldız Teknik Üniversitesi Adresli: Evet

Özet

The NACE-based Q sector, which represents human health and social service activities, has a significant share of employment in Turkey. In the Q sector, the supply of qualified labour process of supplying qualified takes a long time and requires high-cost investments. Due to uncertainties about how and when the demand for health services will arise, the supply should always be higher than the demand. In this context, planning the supply of health services and making predictions about their demands are vital in terms of creating economically effective health service policies. Therefore, this article aims to forecast physicians’ supply and demand for health services. We employed machine learning (ML) methods for time series forecasting. In order to forecast the demand, two data sets from the health care sector between the years 1980–2020 and 2000–2020 were analysed. The supply of physicians per 1,000 people could be 3.04, and the demand 3.12 in 2030, and thus, a shortage in the supply of physicians could be expected in 2030. The main findings of the study demonstrate that there could be an imbalance between the rate at which physicians are expected to be demanded and the rate at which physicians are expected to be supplied.