Of Civil Engineering and dynamics

"Oh...buildings and stuff isn't it. What will you do when you get a job, you will be mixing cement and sand is it? (chuckle), or you will be laying roads? (laughter)."

This is a typical response I get when I tell people that I am pursuing a degree in civil engineering.But when I tell them that I am working on routing and scheduling of feeder-buses to serve the MRTS as a part of my M.Tech project, they are surprised and ask me how this is related to civil engineering?
I tell them that this comes under Transportation Engineering, a sub-branch of civil Engineering.They are even more amused me and ask me, how can that be? Oh, the ignorance is really strong !
So you ask now, what really is civil Engineering? To give a formal definition:
It has to deal with the design, construction, maintenance and operation of Infrastructure. Infrastructure is a very broad term and has to do with electricity supply; basic amenitites like water supply, sewage treatment; buildings and related structures; transportation infrastructure associated with roadways,railways,airports
and ports. People are familiar with the first three components associated with these infrastructure, but it is the operation part that eludes them.
Take for example the transport infrastructure. Roads are built, tracks are laid, who is going to take care of the efficient operation of the transportation on these modes? Traffic policemen? Yeah they can only take care of the short-term measures like fiddling with the signal times,making a road one-way or two-way. Granted, being efficient here would help the cause of smooth movement of traffic. But I say traffic itself is a probelm!! Why have two-wheelers, four-wheelers occupy the roads, when
you can have bus rapid transit systems(BRTS) and Mass rapid transit systems(MRTS) carry more people efficiently at higher speeds(40 -50kmph) than is possible normally(15-20kmph) during peak hour.Ok now we have an MRTS in place in Chennai, but is it serving the purpose? With a daily travel volume of 10, 000 prs/day it hardly qualifies as a MRTS system(which usually carry volumes of the order of 1-2 lakh passengers/day). So what has gone wrong with the MRTS? Why aren't many people using the MRTS?
There are many reasons which can attributed towards this, some of them being it passing through industrial areas, the fares being too high (as compared to the MTC bus system) and that feeder buses are not provided to help the passengers staying in the vicinity of the station to reach it.

It may be difficult to change the first two factors effecting the rider-ship of the MRTS, but the third factor, provision of a feeder-bus system, could be made viable if the routes are designed properly and scheduling of the buses on the so designed routes is done appropriately. The feeder-bus routing problem can be described as follows: There is a station, say Mylapore and there are quite a few people in the vicinity(say 3-5 km radius) of Mylapore willing to use the MRTS from the station, but they don't do so, because there is no bus service provided which they could use to reach the station. Routes need to be designed to cater to the maximum number of such people(demand) with a condition that the total travel time of these routes doesn't exceed a certain limit and the number of buses required to run on these routes also is below the operator's available fleet size and the frequency of buses provided also being below a certain limiting frequency.
And this is where it becomes very exciting, as we are exposed to a variety of fields in solving this problem: the use of operations research in formulating the problem and use of optimisation techniques like dynamic programming, Neural Networks and Genetic algorithms for optimal routing and scheduling, an exposure to the dynamic world of demand modelling, where the travel patterns of passengers are simulated.

An engineer deals with technical systems. When the system is modeled as static, there is not much of uncertainity involved, as is seen in the construction of normal buildings - there is a set formula to deal with such problems. But when dynamic effects need to be taken into account, systems become really exciting to model because the uncertainity associated with the dynamic nature of these systems needs to be taken into account. Hence, to design a high rise building, where the effect of wind loads on the building structure becomes significant, the use of wind-tunnels is made to simulate the effect of high velocity winds on the building.

Non-linear dynamics(NLD) is a well-established field with many real-world applications to weather prediction, fluid flow analysis and any natural flow system analysis.
Non-linear dynamic system or a "choatic" system is one where, the system behaves in a random aperiodic manner. To predict the future behaviour of a chaotic system upto infinity, two things are required:
1) The initial conditions 2) The entire history of data or events since the start of the system. But the second requirement is a serious limitation to the use of chaos theory in these systems. Hence these models are mostly limited to short-term predictions. For example, the meteorological department could forecast the weather maybe upto the next three days and not more than that, because of the second limitation mentioned above.
NLD has a vast potential in the modelling of vehicular transportation and what adds to the complexity is the discreteness of the flow of vehicles or passengers. If NLD is successfully applied to model the behaviour of traffic flow, the implications would be really dramatic and revolutionary, it would provide us with a tool to play around with different parameters and see the effect on the traffic flow patterns.
But the parameters that need to be taken into account are numerous and the time-scale for significant progress in research in this direction is of the order of decades.

All engineering systems, I think, need to be modelled finally as NLD systems or stochastic systems to help simulate the real-world effects as accurately as possible and this is specially the case with transportation system, which is a natural stochastic system and hence to model it is to work on an exciting and a really daunting task.