/Artificial-Intelligence

Electric Vehicle Routing Problem

Primary LanguagePython

Artificial-Intelligence

Problem Statement : Consider a city network where we need to route a set of electric vehicles which may require to be charged during its journey from some source to some destination. Let us assume that we have n cities (v1, v2, . . . , vn) and the distance between cities vi and vj be eij (if two cities are not connected directly then eij = ∞ and eij = eji). Assume that each city has a single charging station which can charge one EV at a time. Consider a set of k EVs namely P1, P2, . . . , Pk. For each EV the following information is provided –

(a) Sr - source node

(b) Dr - destination node

(c) Br - battery charge status initially

(d) cr - charging rate for battery at a charging station (percent charged per unit time)

(e) dr - discharging rate of battery while traveling (distance travel per unit charge)

(f) Mr - maximum battery capacity

(g) sr - average traveling speed (distance per unit time).

Assume that all vehicles start their journey at t = 0 and Pr reaches it destination at t = Tr. We need to route all the vehicles from their respective sources to destinations such that max{Tr} is minimized.

image

Follow the Instructions.txt file for installation and user manual.

Read the Report.txt file for the logic and algorithms used.

Test Case

Node-to-Node Distance

image

Electric Vehicle Details

image

Results obtained

image

The results can be interpreted using the 'Output Interpretation' in the Instructions.txt For the given sample case it does predict the correct results for the given conditions.