Synthesising Activity Participations and Scheduling with Deep Generative Machine Learning

TL;DR

This paper introduces a deep generative machine learning method to synthesize human activity participation and scheduling data, improving speed and simplicity over traditional models while maintaining realism and diversity.

Contribution

The paper presents a novel schedule representation and a comprehensive evaluation framework for deep generative models in activity schedule synthesis.

Findings

The approach can generate large, diverse, and realistic synthetic activity schedules.

It is faster and simpler than existing methods for schedule data synthesis.

The evaluation demonstrates the effectiveness of different schedule encodings and model architectures.

Abstract

Using a deep generative machine learning approach, we synthesise human activity participations and scheduling; i.e. the choices of what activities to participate in and when. Activity schedules are a core component of many applied transport, energy, and epidemiology models. Our data-driven approach directly learns the distributions resulting from human preferences and scheduling logic without the need for complex interacting combinations of sub-models and custom rules. This makes our approach significantly faster and simpler to operate than existing approaches to synthesise or anonymise schedule data. We additionally contribute a novel schedule representation and a comprehensive evaluation framework. We evaluate a range of schedule encoding and deep model architecture combinations. The evaluation shows our approach can rapidly generate large, diverse, novel, and realistic synthetic samples of activity schedules.