Risk and Weightage Based Route Prediction System Using Machine Learning: A Review

Transportation is the key context ingredient in modern life. Optimised route prediction is the vital aspect for commercial enterprises and individuals also. Proper route prediction not only helps to save time, fuel, and money, it additionally prevents accident, air pollution. Operational efficiency like uncertain event and incident can be cause for delay delivery, reaching late in destination that cause also the revenue loss for organisation and individual also. According to the 2019 Urban Mobility Report in 2017, congestion caused urban Americans to travel an extra 8.8 billion hours and purchase an extra 3.3 billion gallons of fuel for a congestion cost of $166 billion. Trucks account for $21 billion (12 percent) of the cost, much more than their 7 percent of traffic. The average auto commuter spends 54 hours in congestion and wastes 21 gallons of fuel due to congestion at a cost of $1,010 in wasted time and fuel. [1]. Nowadays drivers are using different Map software and GPS(Global Positioning System) to choose the appropriate route but sudden changes like climate variation, road congestion due to social gathering, road construction work etc. are difficult to incorporate in existing map software. With the development of Intelligent Transportation Systems (ITS) and Internet of Things (IoT), transportation data has become more and more ubiquitous. This triggers a series of data-driven research to investigate transportation phenomena. Among them, Machine learning theory is considered one of the most promising techniques to tackle tremendous high-dimensional data. Machine Learning models are able to predict more accurate rote in transportation systems. This paper deeply analyses existing route prediction systems and different technology used in those systems. The Authors will focus on improving this route prediction using a machine learning model.