● Lead timetabling and simulation tasks for Ontario’s regional train (GO Train)’s major multi-billion expansion project, enhancing train passenger experiences by reducing delays, shortening total journey times, and increasing train frequencies● Lead a critical task of simulating, developing and implementing a methodology to determine a robust train maintenance schedules, ensuring train operation safety, with highly specialized software● Lead a crucial task of formulating guidelines for train service planning to ensure the train service plans’ compliance to physical and operational constraints of the railway network● Build an interface between planning and simulation software by referencing RailML structure, enabling the seamless transfer of data from planning stage to simulation stage, thereby ensuring the quality and completeness of railway data● Procure major train service planning and simulation software for GO Train’s expansion project; continuously work with software developers to customize functions of the software to suit the ongoing needs of the railway project

• Led a AWS data pipeline upgrade to include real-time data ingestion into Tableau Dashboard of equipment monitoring system with Kinesis, Redshift and Lambda and improved the dashboard response time by 60%• Led the data analytic product team to develop a fault reporting dashboard and reduced escalator faults by 11% by identifying new root causes of faults using chi-square• Estimated the energy consumption of train movement to identify a $2M annual energy-saving opportunities with data mining and a regression model using Python & Tableau• Enhanced the timetabling planning tool of a railway line to improve the operational efficiency by 23% with k-means clustering • Predicted the air-conditioning energy consumption of a building with a stacked tree-based model with 75% accuracy (Extra Tress, XGBoost, Random Forest, LGBM), acting as a basis for future energy optimization• Designed and developed a performance monitoring dashboard along with anomaly detection and trend analyses to save 64 hours of manpower each month using AWS, Python and Tableau• Controlled multiple assigned projects and ensured that expenditures were maintained within the approved budget limitations and can be executed as planned

• Designed and developed data pipelines dashboards and defined KPIs for resource planning visualization using AWS (S3, Glue, Athena, Lambda), Python and Tableau, which saved 24 hours of manual work per month• Designed and implemented the data pipeline for a performance forecasting model supported by regression with >85% accuracy and identified areas for improvements for >$1B-worth projects to enhance the service performance of railway service by >13%• Improved the resource utilization efficiency by 10% and for railway maintenance work planning with data mining and dashboard design and development using Python, AWS and Tableau• Ingest data from disparate data sources using a combination of AWS services, SQL, Python and RPA• Managed data projects with contractors from cradle to grave (preparation of contract documents, tenders, and completion of project work in a timely manner)

• Conducted train service performance analyses and simulations to identify service improvement actions• Developed a performance prediction model to support the introduction schedule of mega projects• Built a travel-time estimation model using linear regression for supporting the proposal of shuttle-bus arrangement under serious railway delay incident to the Commissioner for Transport• Built stochastic models for estimating the confidence level in meeting various operational performance targets• Estimated train availability with Binomial Theorem to support the decision on pursuing additional trains• Facilitated the implementation of a service risk control framework• Ensured RAMS requirements are met for Hung Hum bifurcation design• Technical forum and conference speaker on train service planning with data analytics and optimization algorithms

- Optimisation design of signalling control of an urban network - In cooperation with Transport for London (TfL) and developed a traffic simulation for urban network

Involved in the civil engineering design of the extension of HK MTR Admiralty Station, the busiest underground station in Hong Kong