Matthew J. Berryman’s research while affiliated with University of Wollongong and other places

While aerial shark spotting has been a standard practice for beach safety for decades, new technologies offer enhanced opportunities, ranging from drones/unmanned aerial vehicles (UAVs) that provide new viewing capabilities, to new apps that provide beachgoers with up-to-date risk analysis before entering the water. This report describes the Sharkeye platform, a first-of-its-kind project to demonstrate personal shark alerting for beachgoers in the water and on land, leveraging innovative UAV image collection, cloud-hosted machine learning detection algorithms, and reporting via smart wearables. To execute, our team developed a novel detection algorithm trained via machine learning based on aerial footage of real sharks and rays collected at local beaches, hosted and deployed the algorithm in the cloud, and integrated push alerts to beachgoers in the water via a shark app to run on smartwatches. The project was successfully trialed in the field in Kiama, Australia, with over 350 detection events recorded, followed by the alerting of multiple smartwatches simultaneously both on land and in the water, and with analysis capable of detecting shark analogues, rays, and surfers in average beach conditions, and all based on ~1 h of training data in total. Additional demonstrations showed potential of the system to enable lifeguard-swimmer communication, and the ability to create a network on demand to enable the platform. Our system was developed to provide swimmers and surfers with immediate information via smart apps, empowering lifeguards/lifesavers and beachgoers to prevent unwanted encounters with wildlife before it happens.

Liveability is a concept closely related to the notion of well-being and refers to the environmental conditions that contribute to the quality of life, alongside individual features. Subjective and objective measurements of liveability are both of considerable practical and theoretical importance. A survey is conducted in this paper by which individuals tend to shape their preferences according to six factors describing various aspects of living conditions: (1) home, (2) neighbourhood, (3) transport, (4) entertainment, (5) services and (6) work. The survey data helps us to work out some indicators representing the perceived liveability in the targeted areas. A linear mixed model is used to explore possible relationships between objective factors and perceived liveability. A model-based estimate of liveability index can be then calculated for each non-sampled individual based on his/her socio-demographic characteristics and area of living.

Agent based modelling has been widely accepted as a promising tool for urban planning purposes thanks to its capability to provide sophisticated insights into the social behaviours and the interdependencies that characterise urban systems. In this paper, we report on an agent based model, called TransMob, which explicitly simulates the mutual dynamics between demographic evolution, transport demands, housing needs and the eventual change in the average satisfaction of the residents of an urban area. The ability to reproduce such dynamics is a unique feature that has not been found in many of the like agent based models in the literature. TransMob, is constituted by six major modules: synthetic population, perceived liveability, travel diary assignment, traffic micro-simulator, residential location choice, and travel mode choice. TransMob is used to simulate the dynamics of a metropolitan area in South East of Sydney, Australia, in 2006 and 2011, with demographic evolution. The results are favourably compared against survey data for the area in 2011, therefore validating the capability of TransMob to reproduce the observed complexity of an urban area. We also report on the application of TransMob to simulate various hypothetical scenarios of urban planning policies. We conclude with discussions on current limitations of TransMob, which serve as suggestions for future developments.

Examining travellers' preferences for mode choices to understand a relationship between traveller and transport for New South Wales (TfNSW) from the perspective of agency theory (AT) is the main focus of this paper. This paper emphasises on latent and traditional objective attributes to assess the mode choice process within the agency relationship as indicated in AT as a method by which the utility of the principal (traveller) can be maximised. It is found that the probability of car use is significantly higher than public transport due to mismatch between traveller expectations and present transport services and it indicates an existence of agency problem in this services. Finally, some arguments have been identified to minimise this problem. Thus, the contribution of this research is threefold: firstly, the application of agency theory's utility and implications in traveller choice behaviour; secondly, the demonstration of scale to which attributes influence traveller mode choice to shape the agency relationship within transport mode services; and finally, a pathway for the improvement of agency relationship in transport mode services.

In travel demand models, traditional objective attributes (TOAs) are very commonly used as explanatory variables. Nowadays, it is understood that latent variables (LVs) also significantly influence travellers' behaviour. A hybrid choice modelling approach allows LVs in mode choice utility functions to be addressed. Specifically, a hybrid random parameter logit (HRPL) model has been developed to explore these influences. In this study, a traditional RPL (TRPL) model is compared with an HRPL model. For the later model, a two-step approach (also known as sequential approach) is implemented to incorporate LVs in choice models. Step 1 is the estimation of a MIMIC (multiple indicators and multiple causes) model; a type of regression model with a latent dependent variable(s). Step 2 is the estimation of a choice model with random parameters; information from the first step is incorporated in the second step. The paper analyses and compares the results of applying these models to a real urban case study using two datasets: 2008/09 and 2010/11 household travel survey (HTS) of Sydney Statistical Division (SSD), and also evaluates the predicted changes of mode choice probabilities based on hypothetical scenarios. Our results show that the HRPL model is superior to TRPL models that ignore the effect of LVs on traveller choice. The minimal changes in the parameter coefficients between the two datasets for each model suggest that the changes in traveller choice behaviour are gradual. Three hypothetical scenarios are simulated to forecast the changes that would be relevant to transport policy responses.

Agent based modelling has emerged as a promising tool to provide planners with sophisticated insights on social behaviour and the interdependencies characterising urban systems, particularly with respect to transport and land use planning. This paper presents an agent based simulation, called TransMob, of transport demand and residential mobility in South East Sydney, Australia. In this model, each agent represents an individual resident of the study area. Each agent is given an initial travel diary made of a sequence of daily activities (or trip purposes), associated travel modes and intended departure times. Agents are grouped in various types of household which generate social interdependencies and additional constrains on travel diaries. This initial synthetic population is then allowed to evolve for 20 years, driven by natural growth parameters, social bonding (marriage or divorce) and migration rates. A micro-simulation component calculates daily traffic conditions and individual travel times, allowing for multi-modal journeys. The transport mode choice component uses a multinomial logit model for individual decisions based on various fixed and variable costs as well as socio-economic characteristics. Residential mobility is simulated through a two-step process: first, deciding to move out; then, selecting a destination for relocation. The first step uses a multinomial logit model, while the second one uses a semi-empirical perceived liveability model to inform individual decisions. Simulation results show satisfactory agreement between simulated outputs, socio-demographic changes recorded by the Australian Census and transport patterns recorded by the Household Travel Survey for Sydney Metropolitan Area.