Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Analysis of a new public-transport-service concept: Customized bus in China

In this paper we compare a single reservoir model and a trip-based model under piecewise linear MFD and a piecewise constant demand. These assumptions allow to establish the exact solution of the accumulation-based model, and continuous approximations of the trip-based model at any order using Taylor series.

/pdf/macroscopic-urban-dynamics-analytical-and-numerical-15qa241m75.pdf

Macroscopic urban dynamics: Analytical and numerical comparisons of existing models

Multi-criteria decision analysis framework for sustainable manufacturing in automotive industry

This paper applies the theory of Hamilton–Jacobi partial differential equations to the case of first-order traffic flow models. The traffic flow surface is analyzed with respect to the three 2-dimensional coordinate systems arising in the space of vehicle number, time and distance. In each case, the solution to the initial and boundary value problems are presented. Explicit solution methods and examples are shown for the triangular flow-density diagram case. This unveils new models and shows how a number of existing models are cast as special cases.

The Hamilton–Jacobi partial differential equation and the three representations of traffic flow

This paper identifies a family of linear transformations where conservation laws are invariant. In the case of a triangular fundamental diagram, it is shown that for a subset of these transformations, flow, total distance traveled and total delay are invariant. This means that for capacity or delay computations one may choose the transformation—i.e., the shape of the triangular diagram—that simplifies the problem the most, which does not require knowing the actual fundamental diagram. This is appealing also for delay-optimizing control problems since they may be solved using an isosceles fundamental diagram, which provides the most efficient numerical methods. Examples are given.

/pdf/symmetries-in-the-kinematic-wave-model-and-a-parameter-free-28p2rb3x71.pdf

Symmetries in the kinematic wave model and a parameter-free representation of traffic flow

This paper shows that a wide range of stochastic extensions of the kinematic wave model tend to the same parameter-free expression for the probability of congestion at a given time-space point. This is shown for white noise initial density with deterministic and stochastic fundamental diagram in the case of Riemann problems and the bottleneck problem. It is also found that the stochastic solution i preserves the structure of the deterministic solution and ii tends to the deterministic solution with time at a given location.

The Distribution of Congestion on a Class of Stochastic Kinematic Wave Models

Recent digitalization has led to a significant increase in Data usage and WiFi penetration rates across the globe. WiFi Media Access Control (MAC) data, allows for unannounced and non-participatory tracking of electronic devices. This feature has attracted several transport researchers to use it in the estimation of spatial traffic parameters such as travel time. However, the commercial sensors available for WiFi MAC scanning are not affordable to be installed widely in a city.This paper talks about the development of a portable, low-cost Wireless Media Access Control Scanner and its evaluation for traffic applications. Control and field tests were done to identify the performance of the developed sensor. The developed sensor was found to work comparable or even better in urban traffic scenario when tested against a commercial counterpart.

Development and Evaluation of a low-cost WiFi Media Access Control Scanner as Traffic Sensor

In the absence of pharmaceutical interventions for the Novel Coronavirus (COVID-19), countries have taken drastic steps like quarantine, prohibit large-scale gatherings, limited transport, social distancing, curfews, and lockdowns to curtail the spread of the virus In light of these events, the current study attempts to understand the short-term changes in road traffic patterns, using data from two Wi-Fi MAC Scanners deployed at strategic locations in Chennai, India The results indicate that the road traffic activities significantly reduced due to the restrictions in non-essential trips, workplace suspensions, and strict surveillance during lockdowns However, as the lockdown rules eased, the road traffic activities began to recover It is found that complete closedown is most effective in reducing road travel activity, but ad-hoc short duration complete closedowns may only yield temporary benefits Also, extended lockdowns without proper enforcement may be ineffective since the public appeared to ignore the advisory after a while [ABSTRACT FROM AUTHOR] Copyright of Transportation Letters is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission However, users may print, download, or email articles for individual use This abstract may be abridged No warranty is given about the accuracy of the copy Users should refer to the original published version of the material for the full abstract (Copyright applies to all Abstracts )

https://www.tandfonline.com/doi/pdf/10.1080/19427867.2021.1899580

Analysis of road traffic pattern changes due to activity restrictions during COVID-19 pandemic in Chennai

A transportation infrastructure maintenance management system (TIMMS) for a small urban city is developed. Uintah, Utah, year 2005 population 2,000, is studied. As of Fall 2001, this suburb of the Salt Lake City–Ogden urban area featured 12 km of paved roads, 133 traffic signs, 18 street lights, pavement markings, and other transportation infrastructure. A part-time city engineer was responsible for maintaining the transportation and other infrastructure. A formal TIMMS was not in place. A TIMMS emphasizing preventive maintenance on all infrastructure, along with corrective maintenance on all pavements, was estimated to require about 79% of the city engineer’s time, at an annual cost of about $88,000, or $7,330∕km of road (2002$). About 75% of the expenses would be devoted to pavement preservation. The cost would exceed the city’s estimated transport maintenance budget ($55,775) by about 58%, while the time required of the engineer might exceed that available. A simple linear programming TIMMS formulation...

Bhargava Rama Chilukuri

Papers

Symmetries in the kinematic wave model and a parameter-free representation of traffic flow

The Distribution of Congestion on a Class of Stochastic Kinematic Wave Models

Development and Evaluation of a low-cost WiFi Media Access Control Scanner as Traffic Sensor

Analysis of road traffic pattern changes due to activity restrictions during COVID-19 pandemic in Chennai

Transportation Infrastructure Maintenance Management: Case Study of a Small Urban City