Adaptive multi-horizon models for probabilistic demand forecasting

This project aims to develop an itinerary demand forecasting model that can handle long-term and short-term forecasting and adjust its parameters under changing situations. General long-term prediction models are relatively precise because the context often remains stationary over time, but can not quickly adapt to unforeseen events, like the global pandemics. It is necessary to develop an adaptive model with multi-horizon perspectives. The model will integrate external data sources to output a plausible range of future booking status. With the
understanding and results achieved by this project, accurate and real-time improvement solutions could be proposed and implemented. Therefore, it makes economic sense to understand the customer’s travel behaviors and then adjust the retail practices if unforeseen events occur. This project is expected to produce practical results benefiting the public in improved customer experience, increased incomes, analysis of COVID-19 impacts, etc.

Faculty Supervisor:

Lijun Sun

Student:

Dingyi Zhuang

Partner:

ExPretio Technologies Inc

Discipline:

Engineering - civil

Sector:

Professional, scientific and technical services

University:

McGill University