Performance Measurement and Improvement of Healthcare Service Using Discrete Event Simulation in Bahir Dar Clinic

This paper deals with the service performance analysis and improvement using discrete event simulation has been used. The simulation of the heath care has been done by arena master development 14-version software. The performance measurement for this study are patients output, service rate, service efficiency and it is directly related to waiting time of patients in each service station, work in progress, resource utilization. Simulation model was building for Bahir Dar clinic and then, prepared the proposed model for the system. Based on the simulation model run result, the output of the existing healthcare service system is low due to presence of bottlenecks on the service system. Moreover, the station with the largest queue and high resource utilization are identified as a bottleneck. The bottlenecks, which have identified are reduced by using reassigning the existing resources and add new resources and merging the similar services, which has under low resource utilization (nurses). Finally, the researchers have proposed a developed model from different scenarios. Moreover, the best scenario is developed by combining scenario 2 and 3. And then, service efficiency of the healthcare has increased by 9.86 percent, the work in progress (WIP) are reduced by 3 patients from the system and the service capacity of the system is increased 34 to 40 patients per day due to the reduction of bottleneck stations.


Introduction
Healthcare industry is among the largest industries in Ethiopia.Its service provision is one of the priority service area for the population. In order to acheive this health is becoming a growing concern from time to time in Ethiopia and healthcare centers have been built. According WHO report in 2010 and ministry of Ethiopian health office in 2010 showed that there is 17.7% to 46% growth in the overall health care centers (Eshetie, Selam, & Sisay, 2018). And (Sally & Shivam, 2010) showed that, the quality and responsibility of medical care service in Ethiopia is still among the least ones as compared to the sub Saharan African countries. Heathcare service are confront challengs are physician shift from one place to another, lack of take responsiveness etc.In addition to this, (Eshetie, Selam, & Sisay, 2018) found that designing an optimum balance between customers' demand and available resources in the clinic is one of the main problem has subsequently resulted in long patient waiting time in the healthcare centers. and hospitals in Ethiopia. The amount of time that patients wait to receive service in healthcare centers is one factor, which affects the performances of the health care services. Patient satisfaction is one important quality parameter in healthcare centers. (Kelton, 2002) showed that discrete event simulation has the capability to characterize complex systems and in healthcare can facilitate the decision-making process for operational and management decisions. (Eshetie, Selam, & Sisay, 2018) showed that Simulation studies can optimize patient scheduling, resource utilization, healthcare decision making, patient flow and patient throughput.
Discrete event simulation is widely used in the simulation of healthcare systems, agent technology is a good choice for use in healthcare applications as it best characterizes the operation of complex systems such as emergency department (Eduardo, Manel, MaLuisa, Francisco, & Emilio, 2012).
The advantages of the simulation approach derive from its flexibility, as well as from its ability to manage the variability, uncertainty and complexity of dynamic systems. Simulation is particularly useful when a problem has significant uncertainties, which require stochastic analyzes (MIELCZAREK, 2016).
A clear understanding of this often overlooked concept is crucial for the healthcare model community, which is seeking better stakeholder engagement, demonstration of value and quality assessment (Junqiao, David, Mónica, & Alexandra, 2019).
Bahir Dar health center ( Bahir Dar clinic ) is one of the 10 th heath care facility service found in the city which is established in 1951 E.C. It is located in the capital city of amhara region bahir dar city whichis565 km northern of addis abeba. The facility plans about 60,572 population to serve in ayear.

Statement of the problem
Now adays, Ethiopian healthy sector service delivery system improvment has been practiced by many health center. Howerever, they do not faced their serous challenge that happed in the patients. Long waiting time has found to be a major source of patient dissatisfaction and it increases the proportion of patients who leave without seen by a doctors.
Averagely in Bahir Dar clinic 70 patients are arrived in a day. But they have served 35 patients in a day. So they served only 50% of the total arrived patients.The reason behind the problems are servicing and their resource allocation are not well organized. As a result, the longest patients waiting time is 180 minutes in a system to get a service. Due to this reason, patients had exposed to extra cost, stress and fatigue.
➢ To propose and develop an improved model of the clinic 2. Literature review According to (Haussmann, 1970) waiting time is an important determinate factor of quality services as it is noted that in health care provision delays are expensive and in terms of the potential costs of decreasing patient satisfaction and adverse outcome. (Kelton, 2002) shows simulation refers to the broad collection of application and methods to mimic the characteristics of real systems, usually on a computer with Rockwell arena software. Since many simulation models involve waiting lines or queues as a building block, we would start a very simple case such a model representing a portion of a servicing facility. Patient arrives to a servicing center; they processed by a single channel, and then leave.
Healthcare organizations have come under increasing pressure to provide quality care by addressing rising costs, lower reimbursements and new regulatory demands. Discrete event simulation has become a popular and effective decision-making tool for optimal allocation of scarce healthcare resources to improve patient flow, minimizing healthcare costs and increasing patient satisfaction (Sheldon, Shane, & James, 2006). This is the logical structure of a model: In this document, you can get high-quality solutions in seconds compared to manually prepared schedules that take a lot of time and effort (Topaloglu, 2009 In discrete event simulation model, the variables that describe the system do not change between successive events of the patients (Kelton, 2002).
Independent arrivals and scheduled appointments, as well as new declarations and features, have been carefully designed to solve unique simulation problems specific to hospitals and healthcare (Heflin & Harrell, 1998).  (Vos, Groothus, & Van, 2007) showed that health care organization has been a view with the context of queuing system in which patients arrive, wait for service, obtain service, and then depart from the health center.

Theoretical background of waiting time in health care
The systems engineering design and development processes are examined with particular attention to the discovery of requirements, models and simulation scenarios.

Factors associated with waiting time a health facility
Patient flow: Patient flow shows the ability of the healthcare system to serve patients quickly and efficiently as they move through stages of care.
Blockage in the flow can increase waiting and throughput time creating a negative effect on the quality of service delivery (Vos, Groothus, & Van, 2007).

High Workload:
The physicians are overworked, and then patients have to wait longer as staffs have too many patients to attend. This can be solved by shifting staff from facilities with a low workload.

Lack of efficiency:
Patients may not effectively attend to because much as physicians are present at the service point they are busy with something else; such as teaching, administrative work or preparation.
A logistical problem: Patients may be waiting to see and physician is available to see patients but due to a lack of equipment, rooms or other logistical needs, physician is unable to attend the patients. There was physician present but patients waiting.
Queuing problems: This occurs when patients attended to clinic in an illogical order, that is the patients are not attending in the order that they arrive at the service point.

Methodology
The data and the current information are need for a better and prefect well, since the project is dependent on the current actual data of the case company. The researcher has used some techniques for gathering and acquiring information from our case campany, which is bahir dar clinic.

Sample size determination
Sample size criteria to determine the appropriate sample size: Purpose of the study, the population size, the level of precision, the level of confidence or risk and the degree of variability in the attribute being measured.
The level of precision: Sometimes-called sampling error is the range in which the true value of the population has been estimating. This range is often expressing as in percentage point ±5% (Israel, 1992).  (Israel, 1992). From the previous report of the clinic found that 50% of patients said they had servedin a day. Therefore, we shall use this proportion as p=50%, which means 50% of the patients depart from the system after getting service, but 50% of the patients were not get a service from the system in a day.   The above table shows that, the type of operation, statistical distribution of the operation time and the resource which is assigned in each operation.

Development of standard simulation model
The  Conversely, if there is a statistically significant difference, then the model is not valid and needs additional work before further analysis may conduct.

Model verification and validation
The output of Bahir Dar clinic model in the real service system in eight hours is 50% of patients averagely. The output level the simulation run model in eight-hour is 52% of patients averagely. The output of the simulation model approaches the average output of the real system. Therefore, the model said to be represent the real system and to be valid. (Aregawi, Serajul, & Evan, 2017) and (  The half width statistic used to help in determining the reliability of the results from the replication. In other word, half width is a sampling error in taking sample.

Number of replications
Therefore, the value of half width can be simply n ≅160 …option 2 Therefore, from the two alternatives the maximum number of replication has selected that is 160 (Aregawi, Serajul, & Evan, 2017).

Simulation Model Run Results and Interpretation
The output generated from run simulation model used to know or predict the performance of the system. The output analysis also used to predict the initial model performance and look after the weakness. Therefore, based on the output of the simulation model the performance measures analyzed for the existing service system of the clinic. The performance measures selected to analysis are entity performance, process performance, queue performance and resource performance.
Entity performance: the following points identified from the entity performance: ✓ The WIP of the service is high.

Increase level of resource at bottleneck stations with high WIP
These is focused on the change of the levels of resources at the bottleneck operations and high levels of work in process of the laboratory stations. Adding the resource at the bottleneck operation to reduces the work in progress and then increases the output of the system. Adding one resources for urine test that is more waiting time than other laboratory tests as shown

Combination of alternative scenarios
These alternative scenarios are proposed by combining scenario two and scenario three together.
The simulation run result of combination of alternative scenarios is shown below in the figure.

Comparison of scenarios
The above scenarios simulation run results are compared with the existing simulation model run results. The comparisons of each alternative for the system are as follows The performance measure of the proposed scenario is better than the existing performance measure for the model of the system. From the above proposed scenarios scenario 3, combined scenario 1 and 3 and combined scenario 2 and 3 gives equal performance measure for the model of the system. However, combined scenario 2 and 3 gives better performance measure for the model of the system because this result gives by using low resources than the other scenarios that have the same performance measurements Therefore, the company recommended that to take considerations about combined scenario 2 and 3.

Conclusion
In this paper the service performance analysis and improvement has been analyzing using discrete event simulation. The simulation model of the system has been done by arena master development 14-version software. The performance measurement for this system are output, service rate, service efficiency and it is directly related to waiting time, work in progress, resource utilization.
Simulation model was building for Bahir Dar clinic and then prepared the proposed model for the system.