Activity Sequence Generation Using Universal Mobility Patterns

Abstract

Previous work has established that rank ordered single-day activity sequences from various study areas exhibit a universal power law distribution called Zipf's law. By analyzing datasets from across the world, evidence was provided that it is in fact a universal distribution. This study focuses on a potential mechanism that leads to the power law distribution that was previously discovered. It makes use of 15 household travel survey (HTS) datasets from study areas all over the world to demonstrate that reasonably accurate sets of activity sequences (or ''schedules'') can be generated with extremely little information required; the model requires no input data and contains few tunable parameters. The activity sequence generation mechanism is based on sequential sampling from two universal distributions: (i) the distributions of the number of activities (trips) and (ii) the activity types (trip purposes). This paper also attempts to demonstrate the universal nature of these distributions by fitting several equations to the 15 HTS datasets. The lightweight activity sequence generation model can be implemented in any (lightweight) transportation model to create a basic set of activity sequences, saving effort and cost in data collection and in model development and calibration. Keywords Zipf 's law, activity sequences, universal distributions, trip purpose, number of trips, daily activity pattern Humanity is increasingly challenged with transportation-related issues that have economic, social and ecological consequences. Transportation models are employed to try to find solutions for such problems. They can support ex-ante management decision-making by providing information about the impacts of alternative transportation, land use investments and various other policies, as well as demographic and economic trends. The demand (and need) for such models is steadily increasing. Current models are efficacious, yet they are data-hungry and costly to deploy or transfer to other study areas, making their introduction into all policy decision-making more difficult. Moreover, such models are unattainable for smaller governmental bodies or cities, or whenever scenario results need to be produced quickly. This gives rise to the need for lightweight, easy to deploy modeling solutions. Demand generation, a large component in every modern transport model (be it a four-step model or an activity-based [AB] model), typically requires a significant amount of household travel survey (HTS) data, usually collected via various (expensive and time-consuming) surveying techniques. Other approaches attempt to use mobile phone data (e.g., call detailed records), which could aid data collection but also introduce new challenges (big data processing, inferring of non-observed properties, privacy, etc.). This paper provides insights which may considerably reduce data dependency for some applications. It discusses an activity sequence generation mechanism that is based on sequential sampling of two universal distributions. The model requires no input data and contains almost no tunable parameters. Lightweight demand generation models may be developed, making transportation models more accessible for small government bodies, city managers and so forth. Lightweight models may also