Train Data Analysis for
Mass Transit Railway (MTR)

HKU CS COMP4801 2019-20 Final Year Project

A Large-Scale Industry Project to Develop a Train Data Visualization and Analysis Platform Aiming to Minimize Train Delay through Optimization of Train Schedule and Movement for Incident Response Recommendation

Introduction

This final year project is part of a research project with collaboration between the MTR and the University of Hong Kong (HKU). A team comprising of students and professors from a variety of departments, including but not limited to Computer Science, Statistics and Mathematics has been assembled to provide technical expertise in diverse fields ranging from software development to data analysis.

The goal of this project is to work with the MTR, in order to develop a software platform that analyses and visualises train schedule and movement, for example, the amount of time the train spends at a platform, the departure time of the train, the amount of time for which the train stops in a tunnel etc. The research project aims to reduce the number of trains that suffer from a short delay, i.e. greater than or equal to 5 minutes and long delay, i.e. 30 minutes or above.

MTR System Map

Project Team

Prof. Reynold Cheng

Supervisor

The project is supervised by Prof. Reynold Cheng, Professor of the Department of Computer Science, HKU. Prof. Cheng's main research area is large-scale data management and has been working on the modeling, querying, cleaning, mining, and system development of uncertain databases.

Eashan Trehan

Student

Eashan Trehan is pursuing a Computer Science and Finance double major at HKU and has attended exchange programs at the University of Toronto (UofT) and the London School of Economics and Political Science (LSE), with past work experience at J.P. Morgan and Deutsche Bank.

Aditya Mehta

Student

Aditya Mehta is a Computer Science and Finance double major, with experience in Machine Learning and has previously worked at Hong Kong and Shanghai Banking Corporation (HSBC).

Pang Ming Kin

Student

Pang Ming Kin is a Computer Science student with prior experience in Data Analysis and Web Development, and has previously worked as an intern at the Alibaba Group and China Merchants Bank (CMB).

Research Team

Prof. Michael Ng

Chair Professor

Prof. Michael Ng is Director of Research Division for Mathematical and Statistical Science, and Director of Big Data Research Cluster, Faculty of Science, HKU. His research interests are Applied and Computational Mathematics, Artificial Intelligence and Machine Learning, Data and Imaging Sciences and Scientific Computing.

Prof. Wai-Keung Li

Honorary Professor

Professor LI Wai Keung is Chair of Statistics at The University of Hong Kong. Professor Li is renowned for his works in time series analysis. He has made contributions to almost every area in time-domain time series analysis over the years. In particular, he is an expert in time series model diagnostics. He is also one of the pioneers in considering long memory time series.

Prof. Wai-Ki Ching

Professor

Wai-Ki Ching is a Professor in the Department of Mathematics, the University of Hong Kong. Ching’s research interests are stochastic modeling and matrix computations. In particular, the applications of stochastic models and numerical algorithms in solving problems related to Markov chain and its applications, bioinformatics, image processing, management sciences and quantitative finance.

Prof. Ben Kao

Professor

Professor Kao has published extensively in various international journals, conference proceedings, and books. He has also served as program committee member for many international computer conferences and as reviewer for many journals. His research interests include database management systems, data mining, real-time systems, and information retrieval systems.

Dr. Philip Yu

Associate Professor

Philip Yu is an Assistant Dean (Taught Postgraduate Programmes), the Faculty of Science and an Associate Professor at the Department of Statistics and Actuarial Science of the University of Hong Kong (HKU). His research interests are broad and include non-parametric inference, ranking methods, time series analysis, financial data analysis, risk management and statistical trading.

Mr. Kwan Wing Keung

Assistant IT Director

Mr Kwan is a research computing expert, supporting the research computing services in the Hong Kong University (HKU) during the last more than 30 years. Because of the strong demand in high performance computing (HPC) within HKU, Mr. Kwan has been asked to take charge of the HPC facilities and support in the HKU Computer Centre since 1994.

Wenya Sun

PhD Candidate

Wenya Sun is a PhD candidate of The University of Hong Kong, under the supervision of Dr. Reynold Cheng. Before that, she received my master degree from Shenzhen University in 2019 and Bachelor's degree from the Hebei University of Engineering in 2016. Her research area is Data Mining.

Dr. Tobias Grubenmann

Research Associate

Dr. Tobias Grubenmann is a Postdoctoral Researcher at The University of Hong Kong. He has been a Research Assistant at the University of Zurich from 2014-2018, and a Software Engineer Intern at Google for 4 months in 2016.

Nancy Lam

Research Assistant

Nancy Lam is a Computer Science student with experience in Game Development and has previously worked in Hong Kong Airlines.

Progress

  1. Phase 1
    (2019 Aug - Sep)

    ☑️ Data Collection and Analysis
    ☑️ Simulator Prototype
    ☑️ Project Plan
    ☑️ Project Website
  2. Phase 2
    (2019 Oct - 2020 Jan)

    ☑️ Initial Simulator Implementation
    ☑️ Simulation Model
    ☑️ Result Evaluation
    ☑️ Interim Report
  3. Phase 3
    (2020 Feb - Apr)

    ☑️ Finalized Simulator Implementation
    ☑️ Testing and Validation
    ☑️ Data Preprocessing Pipeline
    ☑️ Final Report

Documentation

Project Plan Cover

Project Plan

Interim Report Cover

Interim Report

Final Report Cover

Final Report