Modeling a Life Queue-Event Data for Selected ATM Service-Point(s)

Abstract

Automatic Teller Machine (ATM) is meant to solve a lot of banking services in today technological world. But the problems its meant to solve in most cases is impeded with a lot of bottlenecks such as network failure, in sufficient funds and long delay in loading the ATM with cash. Meanwhile banking system service is known with First-Come-First-Serve (FCFS) discipline known as queue. Effective queue utilization is based on how efficient is the system in terms of service rate to the number of arrivals per time. The current developments in Rufus Giwa Polytechnic as our case study has also witness deployment of some ATMs within the school. But the incessant and unending long queues being experienced in the ATM point is a matter of serious concern. Dataset of queue records were collected and analyzed. The results show that the available ATM points (2 points) were overloaded, service utilization from each service point was above the threshold, and cases of long waiting time were noticed. While experimenting with R language the mean arrival rate and mean service rate for more than two servers a better performance in the system was observed. To a large extent better service delivery is achievable with 4 to 5 ATM service points with a trade-off of likely idle time to be experience while the customers (most students are on holidays).

Keywords: Automatic Teller Machine (ATM), First-Come-First-Serve (FCFS), Queue, Service utilization, dataset.