Site Loader
Rock Street, San Francisco

AI techniques could be used to solve
transportation problems due to some characteristics that such problems feature

problems often include both qualitative and quantitative data, which allows us
to use AI, say fuzzy systems, to address such issues.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

traditional and rather simple methods normally used for solving transportation
problems often fail to take the complex interactions between the different
parties into consideration. Such complexities arise from the human error uncertainty
and lack of proper understanding of the various interactions.

a transportation problem brings about challenging optimization problems that
cannot be easily solved using the simple mathematical methods, which is where
AI comes into picture.


For over 20 years, KBS has been used to solve
transportation problems. It has basically three components – (a) a knowledge
base in the form of rules, frames or objects, (b) an inference engine in the
form of algorithms on how to process the knowledge and (c) a database, that
will act as the system’s window on the world. The defining feature of KBS as
compared to other software systems is its separation of knowledge base from its
inference engine, so basically the separation of the knowledge about a process
from the process of solving it. This feature enables addition, removal or
modification of knowledge as necessary. In recent years, the focus has shifted
from developing new and independent KBS to integrating them into other
paradigms, like GIS (geographic information systems) or NN (neural networks).

(Transportation Research Board, Artificial Intelligence and Advanced Computing
Applications Committee)Knowledge-based system AI could be used to solve
transportation problems in SCM (supply chain management) like diagnosing
hazardous highway locations, dispatch and control of rail and transit, scheduling
data-driven maintenance activities, planning construction activities and so on.

A simple KBS could be seen below.Genetic Algorithms (GA), a part of the
evolutionary computing, is highly used in . It imitates the principles of
natural evolution and derives a set of rules from natural selection processes
that create objects that most fit the surrounding environment. It has been used
to successfully solve challenging supply chain network design problems like
vehicle routing and scheduling (Park 2001), delivery and pickup (Jung and
Haghani, 2000), bus network optimization (Bielli et al, 2002) etc. One drawback
of GA is that despite its aim to produce global optimum solutions efficiently,
which any supply chain manager prefers over local efficiency, its population
size cannot be infinite.Another AI technique which has been vastly
used in transportation problems is the ant colony optimization algorithm. It
basically studies the good paths through graphs using probability to solve
computational problems, say in transportation.As seen above, there are several AI
techniques currently in use to address various supply chain transportation
design and network. However, as mentioned earlier, these independent systems do
little in terms of supply chain integration, which is something all supply
chain managers are after. While the research on integrating such transportation
problems is highly limited, AI application in transportation is easily visible
with the advent of self-driving cars, thanks to Elon Musk and Uber, who have
introduced such trucks to fight against accidents and promote productivity. Another
benefit is the reduced cost of labor, which entices the major industry players
into adopting AI in its operations. It is estimated that there will be 10
million self-driving vehicles and more than 250 million smart cars on the road
(David Galland, 2017). That is going to be huge in terms of last mile delivery
to customers from a supply chain perspective. In fact, in October 2016, Otto,
an Uber-owned self-driving truck start-up, delivered 50,000 beer cans after
traveling 120 miles at 55 mph in the US.

Another example was the trial of unmanned
delivery vehicles in Greenwich, London, by Ocado and Oxbotica. Each of the
zero-emission CargoPod vans was used to carry up to 128kg of groceries in eight
delivery lockers, and the trial delivered to over one hundred residential
customers.Indian Railway (IR) is trying to modernize
its world’s fourth largest rail network. Through its partnership with multinational
transportation giants Alstom, Siemens and Stadler Bussnang AG, IR is planning
to set up an electric rail coach factory in West Bengal. This could leverage
the use of rail network against roads, of which congestion costs over $10
billion annually to New Delhi only. (Vinay Lohar, 2017) Such infrastructure development
could lay a foundation for AI application in the Indian railway network.

Post Author: admin


I'm Velma!

Would you like to get a custom essay? How about receiving a customized one?

Check it out