Optimizing University Shuttle Buses to Reduce Students’ Waiting Time Using a Discrete Event Simulation Technique

Abstract

Shuttle buses replacing private transports have long been used to reduce traffic congestion in a university campus.  However, to estimate the optimal number of buses taking into account human behaviour using a manual method to provide a better service is very difficult.  This is especially true when the shuttle buses have multiple routes and are significantly affected by unpredictable human behaviour such as their stochastic arrival at the available bus stops.  This paper thus employs modelling and simulation methodologies specifically a Discrete Event Simulation (DES) technique to observe the effects of various configuration of bus number to student waiting time.  For this, a DES model for bus transportation in a university using various available modules in Arena software was first developed.  The simulation model was then fed with relevant data; e.g., the total number of students having classes at certain time periods, the current allocation of the buses for each bus stop, etc.  The model was later used as a test bed for various configurations and analysis of the number of buses and their effects on students’ waiting time including their optimal number.  The simulation results show that the current number of used buses can be reduced to its optimal number while maintaining student waiting time.  This simulation model employs a new approach of using a “transport” module to transport a huge amount of entities to multiple stations as opposed to its traditional usage to only transport a single entity to a particular station.