Data Science and Data-driven Optimisation in Airline and Airport Operations

课程简介

/ Indicative Syllabus Lectures are used to deliver the fundamental knowledge in relation to various aspects of machine learning, data mining, data analytics, data-driven optimisation and artificial intelligence in airline and airport operations (outcomes a to d). Several laboratories will be made available to equip students with the basic knowledge of data mining, soft computing, optimisation and artificial intelligence in solving aviation engineering problems (outcomes a to c). Given the basic knowledge of data science, a group mini project will be used to help students deepen their knowledge of a specific topic through literature study, methodology study, analysis of data, dissemination of research findings and report writing (outcomes a to d). The subject covers the following topics. -- 1 of 4 -- 2 Machine learning, data mining and artificial intelligence - The topics include the following elements:  Supervise and unsupervised learning approach.  Descriptive methods, including clustering, association.  Predictive methods, including classification and regression.  Supervised learning algorithms: Nearest neighbour algorithm, fuzzy logic, gaussian mixture, neural network, linear regression, logistic regression, decision trees, Naïve Bayes, genetic algorithms  Unsupervised learning algorithms: associate rules, principal component analysis, gaussian mixture Data-driven optimisation - The topics include the following elements:  Basic mathematical formulation and modelling, convex optimisation, data-driven modelling, airline scheduling planning, crew rostering, runway scheduling, gate assignment problem, air logistics transportation problem Optimisation methods and soft computing - The topics include the following elements:  Branch and Bound algorithm, heuristics, meta-heuristics, swarm intelligence

目标

This subject will provide students with 1. A conceptual and practical foundation in airport and airline operations for knowledge representation and reasoning of artificial intelligence, data mining, soft computing and optimisation methods as problem solving tools; and 2. Research methodology, data interpretation and analytical skills in regard to real-life data and case scenarios of airport and airline operations; and 3. Experience of conducting proper research experiments and engineering reports for results dissemination.

先修要求

/ Co-requisite/ Exclusion Pre-requisite: AAE3009 Operations Research and Computational Analytics in Air Transport Operations

Teaching Pattern

Methodology Teaching is conducted through class lectures, case studies, and laboratory exercises. The basic knowledge, research methodology and theoretical models will be introduced. The understanding of how to address and formulate problems by using mathematical programming, artificial intelligence algorithms, and soft computing techniques with modern programming language is emphasised. Research methodology, data analytics skills, algorithm design skills and programme methods are taught in class as well as the related real-life scenarios using data to enhance their research abilities. Laboratory exercises, mini reports, oral disseminations and test are used to make up the course work marks.