Decision support system and the adoption of lean in a Swedish emergency ward: Balancing supply and demand towards improved value stream

Facilitated by a decision support system tool, the purpose of this paper is to find the “best” allocated number of surgeons and medicine doctors that reduce patients' non‐value‐added time (NVAT) and total time in the system (TTS).

Interview and observation are first conducted in order to get general insights about (and to understand) the emergency ward of Sahlgrenska Hospital in Gothenburg (Sweden) and its value stream (flow). Then, time‐related data are collected by conducting time measurements empirically and through the triage database. The statistics of the collected empirical data represent the initial state of the system and are utilised as the input of ARENA^® simulation. A simulation scenario is designed by constructing a 3×3 table (= nine combinations) that contains a varying number of surgeons and medicine doctors allocated in the emergency ward. For each combination, 1,000 replications apply (=10 runs @ 100 replications). “Runs” are the cycles or how many times the simulation is executed, while “replications” refer to how many times a computer (automatically) repeats the simulation in a single execution. The simulation length of a single replication was set at 24 hours due to the fact that an emergency ward was always open. The selected feasible solution is the “best” combination of surgeons and medicine doctors that reduces the existing NVAT and TTS while ensuring that the resource utilisation is at a “reasonable” level (and did not exceed 100 per cent).

The simulation output indicates that the emergency ward may achieve considerable reduction in a patients' NVAT and total patients' time in the system by assigning three medicine doctors and three surgeons. This combination leads to (in average) 13 per cent reduction of NVAT while maintaining the TTS at approximately the same level.

An expanded simulation model with a higher level of complexity and ability to accommodate, e.g. cost of care, flow/layout reconfiguration would be greatly needed and is of interest. It would also be relevant to add greater flexibility by assigning more parameters in the simulation model (other than medicine doctor and surgeon).

Simulation can be considered as a valuable decision‐support tool in the adoption of lean in healthcare due to its flexibility in the sense that it is able to show the output (outcome) of various scenarios before any actual change is made. The results of our study present another side of the adoption of lean thinking besides layoff.