Lean Business Services is a government-owned company in Saudi Arabia and the leader in serving and developing innovations for the health sector. The company focuses on digitizing the Saudi health ecosystem and improving healthcare resource utilization.
Problem
Despite extensive emergency coverage, some rural areas still do not have access to all the necessary resources for good public health. In some cases, patients require ambulance transfer to bigger cities where these resources are available. Ambulance transfer, however, has its problems:
- Lack of medical staff
- Long transportation times
- Difficulties for people with serious health conditions
The simulation of medical transfers using AnyLogic as a transportation logistics software was aimed at helping improve patient transportation by tackling ambulance transfer problems. Also, as healthcare analytics software, AnyLogic defined the impact of each major variable and evaluated different scenarios and possibilities, so that decision-makers could develop KPI for each case.
Solution
The proposed solution for patient transportation was to supplement or replace road ambulances with aircraft. Three transfer scenarios were considered when building the model:
- Air transfer between airports only
- Air transfer from hospital to receiving airport
- Air transfer from hospital to hospital
Modeling these scenarios would help study the impact of each scenario compared to the actual ‘as-is’ state of road-only transport.
Lean Business Services developed a hybrid model that employed both agent-based and discrete event modeling. This multimethod approach helped include more information about the scenarios and transportation types than using a single method alone.
Each emergency call is represented as an agent in the model. These agents inherit characteristics from a database containing information about emergency cases, such as patient transport date, sending hospital, receiving hospital, hospital location, etc.
Patients are also represented as agents and have parameters relating to their location, date, and time. The journeys between locations, however, are modeled as discrete event processes because they are best described as a sequence of separate events.
How the agents behave during a simulation run provides insights into the effectiveness of different scenarios. When the model is run and an emergency call is received, results for each transfer scenario are calculated for the KPI and to provide insights. In this way, it is possible to discover the effectiveness of each transfer scenario when faced with many different patient situations.
During an experiment, the model considers various dynamic parameters for better healthcare resource utilization:
- Number of doctors, nurses, ambulance drivers, or other health workers
- Cost of patient transportation
- Night-flight bans
- Short distance air transfer permissions
The simulation model developed by Lean Business Services also contains a GIS map that shows the patient transfer process.
Overall, the company was able to analyze many strategic indicators:
- Average transport duration
- Average number of unavailable medical staff
- Total cost of transportation
- Number of aircraft by type
- Impact of transferring patients (crowding and waiting time)
- Aircraft utilization percentage
- Number of unavailable medical staff per specialty
Result
Based on the outcome of the simulations, the transfer of patients using aircraft for the direct hospital-to-hospital scenario was recommended for three main reasons:
- Low implementation cost
- Lowest average trip time
- Lowest average for the number of unavailable medical staff
Combining discrete event and agent-based simulation paradigms allowed the simulation engineers to make an easily scalable model. It could accommodate the addition of alternative scenarios from transportation logistics software as they were developed, such as changing patient transportation modes, and the addition of new resources at locations.
The case study was presented by Ahmed Alhomaid, of Lean Business Services, at the AnyLogic Conference 2021.
The slides are available as a PDF >>
