Browsing by Author "Dilanthi, M. G. S."

Now showing 1 - 2 of 2

Multi-server queuing system modeling approach for customer service management in a super market: A case study
(4th International Research Symposium on Pure and Applied Sciences, Faculty of Science, University of Kelaniya, Sri Lanka, 2019) Dilanthi, M. G. S.; Ekanayake, H.; Herath, H. M. K. M.; Kaushala, L. M. K.; Ramanayake, R. M. A. M. K.
The service performance is a key determinant for a super market to win customer attraction in large scale. Thus, better performance can be ensured with an efficient service provision to its customers. Accordingly, the study found a renowned supermarket forming long waiting times at its counters and motivated to find solutions for that queuing problem. Therefore, this investigation aimed to analyze the performance of the existing system and provide further improvements to it. This queuing problem was significant there in weekends. Thus the data were collected observing the system from 9.30 am to 12.00 noon in one Saturday. The study recorded customer arrival and service provision times. The sample size was 100 observations. Then the system was modeled and simulated using the student version of Rockwell ARENA 14.5. The study assumed customer arrivals to be random and independent, servers to be identical, no shift changes and breaks for cashiers in the observation period and service discipline to be First In First Out. The data were collected only for three counters considering the recording convenience. Therefore, the system was identified as a multi-server queuing system with infinite waiting room capacity and infinite population. The Input Analyzer showed inter-arrivals to be normally distributed. Also, the service provisions at both counter 1and 2 showed triangular distribution and that of counter 3 showed normal distribution. The system was then modeled using the ARENA modules in both basic process and advanced transfer panels and was run for three hours replication length. The results revealed the service rate of the super market as 76.63%. Furthermore, the average waiting times in the queues at counter 1, 2 and 3 were respectively 16.16, 16.67 and 16.51 minutes and the number waiting were correspondingly 1.62, 1.72 and 1.93. Also, a customer had to spend 21 minutes of average time in the queuing system. Thus, the study made evidence for long waiting times in the queues. Therefore, the super market needed to improve its performance to provide a better satisfaction for their customers. The study emphasized the necessity of trained and efficient cashiers at the counters recommending to provide further training to them and assign helpers to pack goods separately to the counters. In advanced, the cashiers can be facilitated with more sophisticated equipment to do payment transactions.
A Performance Analysis of a Filling Station using ARENA Simulation: A Case Study
(4th International Research Symposium on Pure and Applied Sciences, Faculty of Science, University of Kelaniya, Sri Lanka, 2019) Dilanthi, M. G. S.; Balasooriya, B. M. A. M.; Nimnadi, T. C.; Upekha, T. D. K.; Chathurangani, J. H. D.
The service industry in Sri Lanka is facing the challenge of improving its performance. Thus analyzing its contemporary status is compulsory. This study focused on a petrol filling station in Sri Lanka. The selected system had only one petrol counter with one pump. But it was located in the town area. Therefore, many customers were getting its service, creating a long queue in many times at the filling station. This problem motivated the study to analyze the existing performance with the objective of finding the rate of service provision, average waiting times of customers in the queue and the system and the number of customers waiting in the queue. The study recorded customer arriving and service receiving times as data through direct observation of the system from 2.00 to 4.00 pm of three consecutive weekdays. The sample was 150 observations and both inter arrival times and service times were calculated using them. The system was considered as a single server queuing system with infinite waiting room capacity. Also, the population was infinite. The customers did not leave without being served and the service discipline was identified to be First-Come-First-Served. The study analyzed data using the student version of Rockwell ARENA 14.5. The inter arrival times and service times were then input to the Input Analyzer and obtained Beta and Gamma distributions respectively for them. The system was modeled in ARENA model window with the modules in both basic process and advanced transfer panels. The study ran the model for two hours replication length and gained results. Accordingly, the average service rate of the filling station was 92.6% and average waiting times of a customer in the queue and the system were respectively 4.29 and 5.16 minutes. Also, the average number of vehicles waiting in the queue was six. Therefore, the study recommended to improve the system performance by installing another new service counter with a single pump in the filling station or facilitating the existing counter with another pump considering the economic feasibility. Moreover, a counter with one employee can be located separately to do money transactions and issue the bills. This would also support to minimize the total waiting time of a customer in the filling station