Analysing Road Traffic Situation in Lilongwe: An Agent Based Modelling (ABM) Approach

Abstract

A 15, 451 km road network forms the main mode of transport for Malawi with 26 % paved. With increasing number of vehicles and elongated travel times during rush hour the study analysed the traffic situation on the M1 road between Mchinji and Area 18 roundabouts in Lilongwe City using an agent-based model (ABM). The methodology used game theory’s traffic grip model to analyse traffic flow by controlling traffic variables such as lights, speed limits and the number of vehicles. Each intersection was treated as non-cooperative game where each agent tried to minimize its queue resulting into Q−Nash’s equilibrium as the solution. The ABM tested the empirical relationships of traffic flow parameters in terms of density, flow, acceleration, deceleration, speed, time lost in traffic congestion and fuel consumption. The model was calibrated using traffic data collected through observing 1,312 vehicles sampled against 24,977. The observation results from the road junctions reveal that on average, a vehicle takes 20 mins 18 seconds, 37 minutes 6 seconds, 44 minutes 21 seconds and 58 minutes 53 seconds to exit Chitukuko, Bwandilo, Chilambula roads and Area 18 roundabout respectively upon entering the M1 at Mchinji roundabout. This data was then used to calibrate the business-as-usual model for the peak hour scenario for the road junctions. The model results show that a selected vehicle entering Chitukuko junction travels at an average speed of 22.60 km/hr, until it exits that junction. On average the selected motorists spend 2.52 l/km with a traffic density of 72 v/km. If dualized average speeds improved to 41.54 km/hr while the traffic density declined to 54.42 v/km, saving motorists MK 3,921,624.00 annually. The predictive model of the dual carriage informed that by 2021, commuters will spend MK 5,187,168 on fuel more than single-lane business as usual scenario of 2019.

Game Theory Applications in Micro and Macroscopic Simulations in Transportation Networks: A Comprehensive Review

“…GT can represent a connection with several intersections, with each unit representing a signal regulator that can adapt to changes in the external environment. The complexity arises from the fact that the actions of one operator affect the surroundings and are influenced by the behavior of other operators [109]. Although single agents (players) are self-interested and seek to maximize their gains, this leads to suboptimal junction strategies.…”