Simulation Model Optimizes Community Pharmacy Workflow

Researchers in Japan have developed a discrete-event simulation model to evaluate how operational changes in community pharmacies affect patient total stay time and the time available for medication counseling. The study, published in Discover Health Systems, utilized observational data from a single community pharmacy in Hamamatsu City to construct a digital replica of the pharmacy’s workflow. By simulating various staffing configurations and automation scenarios, the team identified strategies that could reduce patient wait times while simultaneously increasing the duration of pharmacist-led counseling, a key component of patient-centered care.

The investigation was motivated by the evolving role of community pharmacies in Japan, where an aging population and a relative shortage of qualified pharmacists have created pressure to enhance service efficiency. Traditional pharmacy operations in Japan focus heavily on dispensing tasks, including prescription acceptance, verification, medication dispensing, and documentation. However, national health policies, such as the 2015 Vision of Pharmacies for Patients, aim to transform these facilities into patient-centered hubs that provide continuous medication management and support for home healthcare. Despite these goals, current practices often limit medication counseling to brief interactions, frequently under five minutes, which may be insufficient for preventing drug-related problems and ensuring patient adherence.

To address these operational challenges, the study team conducted a five-day observational survey at a community pharmacy adjacent to a general hospital. The pharmacy, selected for its representative medium-sized scale and diverse patient population, operated from 8:45 AM to 6:30 PM with seven pharmacists on staff at all times. The workflow involved a sequential process: patients arrived with prescriptions, which were verified and sent to a second floor for dispensing and auditing by independent pharmacists. After the audit, medications were returned to the first floor for counseling, followed by payment and departure. This sequential nature of the workflow made it particularly suitable for analysis using discrete-event simulation, a computer modeling technique that replicates events occurring at discrete times to predict system behavior.

The researchers collected data from 652 patient visits, with 638 complete datasets included in the analysis. Pharmacist task times were self-reported, while patient demographics and prescription details were obtained from pharmacy records. The average number of prescriptions per day was 146.4, with peak arrival times occurring between 10:00 AM and 12:00 PM and again between 3:00 PM and 4:00 PM. The mean patient total stay time, defined as the duration from reception to payment completion, was 16.0 minutes in the observed data. Notably, patients receiving unit-dose packaging, a common practice for elderly patients with multiple medications, experienced longer stay times of 24.3 minutes compared to 15.2 minutes for those without such packaging.

Using the Simul8 Professional Software package, the team built a baseline model calibrated to match the observed operational data. The model incorporated probability distributions for dispensing, auditing, and counseling times, estimated using StatFit software to reflect real-world variability. Validation of the model showed that the simulated patient total stay time differed from observed data by less than four minutes, confirming its accuracy. This validated baseline served as the reference point for testing four distinct optimization scenarios: adjusting pharmacist staffing levels, reducing dispensing time through assumed automation, increasing medication counseling time, and combining these interventions.

In the first scenario, the researchers simulated changes in pharmacist allocation between the first floor (counseling) and second floor (dispensing and auditing). The baseline configuration had four pharmacists on the first floor and three on the second. Shifting the configuration to three on the first floor and four on the second reduced the simulated patient total stay time from 18.9 minutes to 14.4 minutes. This reallocation also increased the proportion of total stay time dedicated to medication counseling from 21.0% to 27.6%. The results suggested that the second floor was a bottleneck in the original configuration, and moving staff to this area alleviated the congestion, allowing for more efficient processing of prescriptions.

The second scenario explored the impact of dispensing automation, assuming that automated equipment could reduce dispensing time to 70% of its current duration. When combined with the optimized staffing configuration of three pharmacists on the first floor and four on the second, the simulated patient total stay time decreased further to 11.8 minutes. In this scenario, the proportion of time spent on medication counseling rose to 33.6%. The study noted that while automation reduced second-floor processing time, it could paradoxically create a bottleneck on the first floor if staffing was not adjusted accordingly. The configuration of three on the first floor and four on the second proved most efficient, balancing the load between the two floors.

The third scenario examined the effects of extending medication counseling time. Under the baseline staffing, increasing counseling time to three times the current duration extended the patient total stay time to 28.8 minutes. However, with the optimized staffing configuration, the model predicted that counseling time could be extended by up to 1.5 times without exceeding the baseline stay time. This indicated that operational efficiency gains from staffing adjustments could be leveraged to provide more thorough patient education without penalizing patients with longer waits. The fourth scenario combined all previous interventions, demonstrating that the combination of optimized staffing and dispensing automation allowed for counseling time to be extended by up to twice the current duration while keeping patient total stay time at or below baseline levels.

The study highlights the utility of discrete-event simulation as a risk-free environment for testing operational changes before implementation. In healthcare settings, real-world experiments can be costly and expose patients to potential risks, making simulation a valuable tool for evidence-based decision-making. The findings suggest that community pharmacies can enhance patient-centered services by reallocating pharmacist resources and considering automation, particularly as the demand for unit-dose packaging grows with the aging population. The model also revealed that the optimal configuration was consistently three pharmacists on the first floor and four on the second, a finding that could inform workforce planning in similar settings.

Limitations of the study include its reliance on data from a single pharmacy over a short five-day period, which may limit the generalizability of the findings to all community pharmacies in Japan or other countries. The self-reported nature of pharmacist task times may have introduced measurement bias, and the model did not account for the roles of non-pharmacist staff, such as administrative personnel. Additionally, the study did not incorporate the potential for pharmacy technicians, a profession that is well-established in many other countries but not yet formally recognized in Japan. Future research should expand this approach to multiple pharmacies with different operational environments and consider evolving workforce structures to validate these findings and adapt simulation approaches to changing societal needs.

Despite these limitations, the study demonstrates that computer simulation can quantitatively predict changes in pharmacy operations, offering a practical method for improving efficiency and quality of care. By identifying bottlenecks and testing potential solutions, pharmacies can make informed decisions about staffing and technology investments. The results support the global trend of transforming community pharmacies into patient-centered facilities, where pharmacists can devote more time to counseling and patient education. As healthcare systems continue to evolve, tools like discrete-event simulation will play an increasingly important role in optimizing resource allocation and enhancing patient outcomes.

Subject of Research: Optimization of pharmacist operations in community pharmacies using a discrete event simulation model

Article Title: Optimization of pharmacist operations in community pharmacies using a discrete event simulation model

Article References: Furushima, D., Makita, A., Tsuji, D., Watanabe, S., & Yamada, H. (2026). Optimization of pharmacist operations in community pharmacies using a discrete event simulation model. Discover Health Systems, 5(1), Article 63. https://doi.org/10.1007/s44250-026-00353-8

Image Credits: AI Generated

DOI: 10.1007/s44250-026-00353-8

Keywords: Optimization, pharmacist, operations, community, pharmacies, discrete, event, simulation, model, scientific research