EXPLORING STOCHASTIC MODELING AND SYSTEM DYNAMICS FOR HEALTHCARE OPTIMIZATION: LESSONS FOR RADIOTHERAPY SCHEDULING

Abstract

Background: All Radiotherapy schedules are designed to maximise tumour control probability and minimise normal healthy tissue complications. However, the rapidly growing demand for radiotherapy services, coupled with limited resources, poorly scheduled appointments, and increasing complexities of cases, is urgently demanding more advanced systems and models to optimise the efficiency of radiotherapy treatments around the world. This is further complicated by unplanned gaps leading to significant scheduling challenges.

Materials and Methods: This paper focuses on stochastic processes, system dynamics, and artificial intelligence (AI) for radiotherapy scheduling. It utilises a systematic literature review concerning scheduling methodologies that embrace Markov chains, Monte Carlo methods, queuing systems, and their hybrids. The review sought literature to control patient flow, resource control, and adaptable systems.

Results: Stochastic models were found to promote better utilisation of resources by improving patient flow, reducing machine idle time, and enhancing resource utilisation. Adaptive scheduling approaches and system dynamic models enable better workforce planning, reduce idle time, and minimise the disruptions of various business processes. AI-info trick analytics further enhance scheduling accuracy by making arrivals and equipment downtime forecasts to refine resource scheduling approaches. Stochastic approaches to system dynamic modelling offer a holistic way to solve practical scheduling problems.

Conclusion: Advanced scheduling techniques, especially hybrid ones, can enhance resource allocation and reduce treatment delays for radiotherapy patients. Future studies should focus on integrating electronic health records (EHRs) with active patient monitoring to augment scheduling accuracy. Such approaches will increase treatment and service delivery efficacy in radiotherapy centres through a real-time patient data-capturing model while ensuring that all patients receive equally effective treatment.