92 results found
Luan J, Daina N, Reinau KH, et al., 2022, A data-based opportunity identification engine for collaborative freight logistics based on a trailer capacity graph, Expert Systems with Applications, Vol: 210, ISSN: 0957-4174
Logistics operators participating in horizontal collaboration can gain economic benefits and being better placed to meet environmental goals. Data-based approaches provide a viable, albeit suboptimal, solution that can enable real-time collaborative order sharing. Conventional data-based approaches for identifying collaboration (order sharing) opportunities are typically based on origin-destination (OD) matching between trips and shipments from different collaborating companies. This, however, prevents the exploitation of en-route collaboration opportunities. Hence, we propose a practical data-based engine for identifying collaboration opportunities during shipment planning stages that enables shipments to be matched according to both the OD and trailer trip routes. The engine is based on a multigraph approach, called the trailer capacity graph (TCG) approach. We further enhance the engine to improve its computational performance for real-time operations. Numerical experiments based on real-world data from two logistics companies show that the TCG approach identifies a significantly larger number of opportunities, and provides a higher total distance saving than conventional OD-based matching. The experiments also demonstrate that with trailer route approximation and route shape simplification, this engine allows trade-offs between the computational performance and the effectiveness of opportunity identification, which implies that the engine can be flexibly tailored according to user preferences.
Abeille A, Pawlak J, Sivakumar A, 2022, Exploring the meaning and drivers of personal (Un-)Productivity of knowledge workers in mobile settings, Travel Behaviour and Society, Vol: 27, Pages: 26-37, ISSN: 2214-367X
It is now recognised in travel behaviour research that travel time can be, and in fact is, used to undertake productive, work-related activities. The phenomenon, also referred to as travel-based multitasking, has in recent years been compounded by the proliferation and sophistication of mobile information and communication technologies (ICT). Accordingly, several research efforts have made attempts to measure and model the effectiveness of work activities during travel. Yet reliance of those studies on rather simple and proxy metrics has led to a limited understanding of mobile work productivity. This has been especially the case for knowledge workers, whose job involves handling or using information and is often characterized by intangible work outputs. To address this shortcoming, the current paper presents a systematic analysis of 22 semi-structured interviews of employees of a major IT company regarding their mobile work practices, use of ICT and perception of productivity with use of ICT. Analysis of the interviews led us to adopt an ‘inverse’ approach, i.e. discussing factors hampering productivity. This emerged from our observation that individuals experienced difficulties speaking about productivity and productive tasks while finding it easier to discuss what made them unproductive. With the lens of what we term ‘unproductivity’, we are able to provide a new perspective on how to characterize the impacts of journey, technology and individual factors on productivity during episodes of mobile work. In addition, we find a strong link between productive mobile work, planning of the journey and the working activities during the course of travel.
Wu C, Le Vine S, Sivakumar A, et al., 2022, Dynamic pricing of free-floating carsharing networks with sensitivity to travellers’ attitudes towards risk, Transportation, Vol: 49, Pages: 679-702, ISSN: 0049-4488
Free-floating carsharing (FFCS) systems are characterised by volatile fleet distribution as well as customers’ heterogeneous price sensitivity and spatiotemporal flexibility. There is thus an opportunity for operators to employ dynamic pricing to manage various aspects of fleet allocation: which customer is provided which vehicle, at what time and price, and the agreed pick-up and drop-off location. While there are emerging examples of dynamic pricing in FFCS, there is as yet no general framework for the interaction of consumer and operator behaviours in this context, most particularly consumer response to the inherent risks and uncertainties in the journey characteristics noted above. In this study, we propose a choice-based framework for modelling the supply/demand interaction, drawing on behavioural models of decision-making in risky choice contexts and empirical stated-choice data of user preferences in a dynamically priced FFCS market. In addition to the ‘spot market’ mechanism of dynamic pricing, the proposed framework is capable of evaluating operator strategies of allowing (at an agreed price) customers to make guaranteed advance reservations. We demonstrate that this approach allows the system operator to set an optimal pricing strategy regardless of whether user risk preferences are risk-seeking or risk-averse. We also demonstrate the applicability of the proposed framework when the operator seeks to maximise revenue (as with a private operator) vs social welfare (as with a public operator). In the case study which employs empirical user preferences, we show that users’ risk preferences have a relatively small impact on revenue, however the impacts are much larger if there is a mismatch between users’ actual risk preferences and the system operator’s assumptions regarding users’ risk preferences.
Manca F, Sivakumar A, Polak J, 2022, Capturing the effect of multiple social influence sources on the adoption of new transport technologies and services, Journal of Choice Modelling, Vol: 42, Pages: 1-18, ISSN: 1755-5345
This paper presents a conceptual and modelling framework that makes it possible to disentangle and quantify multiple social influence effects affecting the individual’s choice behaviour. The proposed structure simultaneously accounts for live social interaction effects, social influence processes of diffusion, translation and reflexivity, conformity processes related to social norms, such as the hypothetical adoption rate within a social network, and correlated effects related to psychometric attitudinal characteristics of peers. The modelling framework is applied to investigate the adoption of bike-sharing in a student cohort during a public transport strike. A joint hybrid choice model is estimated using a two-wave stated preference dataset and incorporating latent variables, social influence measures and live social interactions with dynamic ‘inertia’ processes. In this empirical context, results and sensitivity analyses show that the social influence variables are highly significant and explain part of the heterogeneity in choosing bike-sharing. A greater utility for this travel mode is associated with a greater hypothetical bike-sharing adoption and with live social interactions improving the understanding of the bike-sharing benefits. The results also suggest that conformity processes and social interaction effects can have a higher impact on the choice than correlated effects related to the attitudinal ‘propensity towards cars’ in the social network. This study, therefore, provides further evidence that, in the context of new technology adoptions, the choice is not only driven by explanatory variables that can be generally observed but may well be affected by social factors facilitating the exchange of information and the understanding of the individuals.
Manca F, Daina N, Sivakumar A, et al., 2022, Using digital social market applications to incentivise active travel: Empirical analysis of a smart city initiative, Sustainable Cities and Society, Vol: 77, ISSN: 2210-6707
Information and communication technologies (ICTs), such as mobile communication networks, and behaviour-based approaches for citizen engagement play a key role in making future cities sustainable and tackling persistent problems in high-density urban areas. In the context of Sharing Cities, an EU-funded programme aiming to deliver smart city solutions in areas such as citizen participation and infrastructure improvements of buildings and mobility, a prominent intervention has been the deployment and monitoring of a Digital Social Market (DSM) tool in Milan (Italy). The DSM allows cities to engage with residents and encourage sustainable behaviours by offering non-monetary rewards. This paper aims to evaluate the effectiveness of the DSM approach to promote active travel (cycling and walking) by analysing the data collected through the app as well as through participant surveys. Our model results show that a broader engagement with the DSM app (number of claps to posts, number of posts made, non-monetary rewards earned by participating in non-travel events) is positively correlated with the monitored level of active travel. Lifestyles, attitudes, and social influence also explain the variability in cycling and walking. This highlights the importance of investigating these factors when replicating such initiatives on a large scale.
Losa Rovira Y, Faghih Imani A, Sivakumar A, et al., 2022, Do in-home and virtual activities impact out-of-home activity participation? Investigating end-user activity behaviour and time use for residential energy applications, Energy and Buildings, Vol: 257, Pages: 1-11, ISSN: 0378-7788
The ability to accurately model and predict timing and duration of activities for different individuals is essential for successful and widespread Demand Side Response (DSR) policies, especially in the residential sector. Understanding what people do during the day and what factors influence their activity participation decisions is important for planning an effective DSR strategy to harness the end-user flexibility. The recent Covid-19 pandemic has shown how much activities can be shifted to a virtual mode in the presence of mobility restrictions. Further, participation in activities via digital devices (virtual activity participation) has spread across society. Such virtual activities, including teleworking, online shopping, and virtual social interactions, are observed to explicitly impact travel behaviour and activity scheduling. And yet, activity-based models of mobility and energy demand do not accommodate the trade-offs between activity types, location and virtual activity participation. This paper presents a model of activity participation that captures the relationship between the three dimensions of: activity type (such as work, study, shopping), activity location (in-home, out-of-home), and activity modality (in-person, virtual). A Multiple-Discrete Continuous Extreme Value (MDCEV) model structure is applied, and the empirical analysis is undertaken using the 2015 United Kingdom Time Use Survey (UKTUS). The model results provide insights for better understanding of the trade-offs made by individuals as they participate in and allocate time across a set of activity type-location-modality alternatives, and the heterogeneity in these trade-offs. Further, holdout sample validation and policy scenario analysis exercises are presented to demonstrate the reliability and suitability of the model for policy implications. The empirical results presented in our paper suggest that this framework embedded in an activity and agent-based simulator of energy demand will
Zhao Y, Pawlak J, Sivakumar A, 2022, Theory for socio-demographic enrichment performance using the inverse discrete choice modelling approach, Transportation Research Part B: Methodological: an international journal, Vol: 155, Pages: 101-134, ISSN: 0191-2615
In light of the growing availability of big data sources and the essential role of socio-demographic information in travel behaviour and transport demand modelling more broadly, the enrichment of socio-demographic attributes for anonymous big datasets is a key issue that continues to be explored. The common shortcoming of existing socio-demographic enrichment approaches concerns their lack of consistent theory that can link their enrichment performance (i.e. the ability to correctly enrich the required attribute) to the underlying covariance structure in the anonymous big datasets. In other words, existing approaches are unable to indicate, prior to the enrichment, to what extent it will be successful. Instead, they require undertaking the enrichment itself to assess and validate it post factum, incurring the effort and cost of the activity. An alternative and arguably preferable way would be to have a prior indicator as to whether an enrichment is likely to be sufficiently effective for the desired application.Towards this end, this paper draws upon the Inverse Discrete Choice Modelling (IDCM) approach to demonstrate what is termed as the IDCM performance theory, which systematically and in a tractable manner links the socio-demographic enrichment performance of the IDCM approach to the structure of the underlying datasets. This is achieved by recalibration of the constant, a technique adopted from conventional discrete choice modelling practice, while also drawing upon information theory employed in the context of communication systems. The established IDCM performance theory is validated in two empirical applications where performance of the IDCM approach in enriching several socio-demographic attributes, given travel behaviour patterns, is successfully estimated. Additionally, the IDCM approach is found to perform comparably to commonly used methods in previous socio-demographic enrichment efforts. It is thus argued that the capability of the IDCM performance th
Pawlak J, Faghih Imani A, Sivakumar A, 2021, How do household activities drive electricity demand? Applying activity-based modelling in the context of the United Kingdom, Energy Research and Social Science, Vol: 82, Pages: 1-18, ISSN: 2214-6296
Driven by the necessity to increase utilisation of the existing networks and accommodation of volatility in renewable energy generation, the energy sector is undergoing a shift from an unconstrained infrastructure expansion to accommodate growth in demand towards demand management strategies. Such strategies, for example nudging demand using incentives such as price signals, or Demand Side Response (DSR), rely on the ability to accurately understand and harness flexibility in demand. Activity-based demand modelling frameworks can provide this capability, as they enable the detailed modelling and simulation of individuals and their activities. However, to date, no modelling approach has been proposed that can link energy consumption of a household to the activities undertaken, heterogeneity of the household residents, presence and use of household appliances and devices as well as weather and energy system-related variables. This paper addresses the gap by proposing a log-linear mixed-effects model of energy consumption based on reported household activities alongside a comprehensive set of attributes and contextual variables that might influence household energy consumption. Application of the model is demonstrated using joint time-use and residential electricity consumption data from 160 households, collected between 2016 and 2018 in the UK. The modelling results prove the value of incorporating time-use (activities) in modelling residentialelectricity demand, when compared against modelling without such considerations. Furthermore, the model provides (semi-)elasticities of demand and marginal changes in electricity consumption due to activities, which are of direct policy value or serve as inputs into activity-based energy demand simulation.
Latinopoulos C, Patrier A, Sivakumar A, 2021, Planning for e-scooter use in metropolitan cities: A case study for Paris, Transportation Research Part D: Transport and Environment, Vol: 100, Pages: 103037-103037, ISSN: 1361-9209
This Briefing Paper explores the impactthe COVID-19 pandemic had on the UK’senergy sector over the course of thefirst government-mandated nationallockdown that began on 23 March 2020.Research from several aspects of theIntegrated Development of Low-carbonEnergy Systems (IDLES) programme atImperial College London is presented inone overarching paper. The main aim isto determine what lessons can be learntfrom that lockdown period, given theunique set of challenges it presented inour daily lives and the changes it broughtabout in energy demand, supply, anduse. Valuable insights are gained intohow working-from-home policies,electric vehicles, and low-carbon gridscan be implemented, incentivised, andmanaged effectively.
Manca F, Sivakumar A, Pawlak J, et al., 2021, Will we fly again? modeling air travel demand in light of COVID-19 through a London case study, Transportation Research Record: Journal of the Transportation Research Board, Pages: 1-13, ISSN: 0361-1981
The COVID-19 pandemic and associated travel restrictions have created an unprecedented challenge for the air transport industry, which before the pandemic was facing almost the exact opposite set of problems. Instead of the growing demand and need for capacity expansion warring against environmental concerns, the sector is now facing a slump in demand and the continuing uncertainty about the impacts of the pandemic on people’s willingness to fly. To shed light on consumer attitudes toward air travel during and post the pandemic, this study presents an analysis that draws on recently collected survey data (April–July 2020), including both revealed and stated preference components, of 388 respondents who traveled from one of the six London, U.K., airports in 2019. Several travel scenarios considering the circumstances and attitudes related to COVID-19 are explored. The data is analyzed using a hybrid choice model to integrate latent constructs related to attitudinal characteristics. The analysis confirms the impact of consumers’ health concerns on their willingness to travel, as a function of travel characteristics, that is, cost and number of transfers. It also provides insights into preference heterogeneity as a function of sociodemographic characteristics. However, no significant effects are observed concerning perceptions of safety arising from wearing a mask, or concerns over the necessity to quarantine. Results also suggest that some respondents may perceive virtual substitutes for business travel, for example video calls and similar software, as only a temporary measure, and seek to return to traveling as soon as it is possible to do so safely.The ongoing COVID-19 pandemic has affected air travel to an unprecedented extent, leading to the worst-ever crisis of the air transport sector (1). Airlines worldwide have faced a huge drop in demand, for example 98% drop in passengers for 6 weeks in a row over April and May 2020, as stated by the Airpor
Li J, Guo F, Sivakumar A, et al., 2021, Transferability Improvement in Short-term Traffic Prediction using Stacked LSTM Network, Transportation Research Part C: Emerging Technologies, Vol: 124, ISSN: 0968-090X
Short-term traffic flow forecasting is a key element in Intelligent Transport Systems (ITS) to provide proactive traffic state information to road network operators. A variety of methods to predict traffic variables in the short-term can be found in the literature, ranging from time-series algorithms, machine learning tools and deep learning methods to a selective hybrid of these approaches. Despite the advances in prediction techniques, a challenging problem that affects the application of such methods in the real world is the prevalence of insufficient data across an entire network. It is rare that extensive historical training data required for model training are available for all the links in a city. In order to address this data insufficiency problem, this paper applies transfer learning techniques to machine learning methods in short-term traffic prediction. All the traffic data used in this paper were collected from Highways England road networks in the UK. The results show that through improving the transferability of machine learning-based models, the computational burden due to the model training process can be significantly reduced and the prediction accuracy under data deficient scenarios can be improved for one-step ahead prediction. However, the prediction accuracy gradually decreases in multi-step ahead prediction. It is also found that the accuracy of the proposed hybrid method is highly dependent upon consistency between datasets but less dependent on geographical attributes of links.
Latinopoulos C, Sivakumar A, Polak J, 2021, Optimal pricing of vehicle-to-grid services using disaggregate demand models, Energies, Vol: 14, ISSN: 1996-1073
The recent revolution in electric mobility is both crucial and promising in the coordinated effort to reduce global emissions and tackle climate change. However, mass electrification brings up new technical problems that need to be solved. The increasing penetration rates of electric vehicles will add an unprecedented energy load to existing power grids. The stability and the quality of power systems, especially on a local distribution level, will be compromised by multiple vehicles that are simultaneously connected to the grid. In this paper, the authors propose a choice-based pricing algorithm to indirectly control the charging and V2G activities of electric vehicles in non-residential facilities. Two metaheuristic approaches were applied to solve the optimization problem, and a comparative analysis was performed to evaluate their performance. The proposed algorithm would result in a significant revenue increase for the parking operator, and at the same time, it could alleviate the overloading of local distribution transformers and postpone heavy infrastructure investments.
Manca F, Sivakumar A, Pawlak J, et al., 2021, Will We Fly Again? Modelling Air Travel Demand in Light of COVID-19 through a London Case Study, 100th Transportation Research Board Annual Meeting
Dong Y, Guo F, Sivakumar A, et al., 2021, Short-term traffic prediction under disruptions using deep learning, Traffic Information and Control, Pages: 79-114, ISBN: 9781839530258
Manca F, Sivakumar A, 2020, Modelling live social interactions and indirect social influence affecting the adoption of transport technologies and services: a bike sharing case study, hEART 2020: 9th Symposium of the European Association for Research in Transportation
Wu C, Le Vine S, Sivakumar A, 2020, Exploratory analysis of young adults' trajectories through the UK driving license acquisition process., Traffic Injury Prevention, Vol: 22, Pages: 37-42, ISSN: 1538-957X
OBJECTIVE: The UK is one of many high-income countries to experience a decline in driving license acquisition among young adults in the 2000s. This paper draws on newly available nationally representative microdata that captures the progress of individual drivers through the UK driving license acquisition process, to establish socio-demographic correlates. METHODS: Using the 2016 and 2017 editions of England's National Travel Survey data, a series of binary logit models were employed to identify factors associated with progression through the various phases of the UK's driving license acquisition process. Factors that are associated with (1) the frequency of taking the driving license tests, (2) the number of times having failed the theory and driving tests are then identified. RESULTS: The socio-demographic explanators considered were each found to be associated with driving license holding in intuitive ways that are consistent with prior literature. However, relatively few factors are significantly associated with progress through the steps of the license acquisition process, and the goodness-of-fit for progress through these intermediate phases are generally lower (indicating that other unobservable idiosyncratic personal or contextual characteristics are dominant in these processes). A consistent theme is the strong relationship with labor market participation. Links between income and the intermediate phases, however, were generally weaker. Age is negatively associated with progress through the early phases when respondents are applying for provisional license and taking theory test, but this relationship turns positive in later stages of the acquisition process. CONCLUSION: To the authors' knowledge, this study is the first opportunity to evaluate this novel data resource covering the UK's driving license acquisition process. This is an important research direction to help policymakers understand young adults' delay in acquiring licenses, particularly the exte
Dixit M, Sivakumar A, 2020, Capturing the impact of individual characteristics on transport accessibility and equity analysis, Transportation Research Part D: Transport and Environment, Vol: 87, Pages: 1-17, ISSN: 1361-9209
Transport accessibility experienced by an individual depends on their needs and abilities, as represented by their individual characteristics, such as age, income and gender. Although important from an equity perspective, the individual component of accessibility is currently ignored in most transport equity studies.This paper evaluates the impact of including individual characteristics into logsum-based accessibility measures for transport equity analysis. Using data from the London Travel Demand Survey (LTDS) 2011-13, two alternate log sum measures of accessibility are specified –with and without individual characteristics. An empirical analysis of spatial, social and economic equity is conducted using both the measures,and the outcomes are compared.The results clearly demonstrate that ignoring individual characteristics in logsum measures of accessibility can lead to unreliable outcomes for social and economic equity analysis,but do not add significant value when aggregated across large geographical zones for spatial equity analysis.Overall, ignoring individual characteristics masks the disparity in distribution of accessibility, as measured by the Gini index. Although not straightforward, the difference between accessibility patterns using the two logsum measures also yields insights into the possible causes of inequity, which can provide actionable inputs to policy makers.The study highlights that personal needs and abilities are often responsible for accessibility variations among individuals and ignoring them can result in a misleading picture of equity,as demonstrated quantitatively in this paper.
Hou H, Pawlak J, Sivakumar A, et al., 2020, An approach for building occupancy modelling considering the urban context, Building and Environment, Vol: 183, Pages: 1-18, ISSN: 0360-1323
Building occupancy, which reflects occupant presence, movements and activities within the building space, is a key factor to consider in building energy modelling and simulation. Characterising complex occupant behaviours and their determinants poses challenges from the sensing, modelling, interpretation and prediction perspectives. Past studies typically applied time-dependent models to predict regular occupancy patterns for commercial buildings. However, this prevalent reliance on purely time-of-day effects is typically not sufficient to accurately characterise the complex occupancy patterns as they may vary with building’s surrounding conditions, i.e. the urban environment. Therefore, this research proposes a conceptual framework to incorporate the interactions between urban systems and building occupancy. Under the framework, we propose a novel modelling methodology relying on competing risk hazard formulation to analyse the occupancy of a case study building in London, UK. The occupancy profiles were inferred from the Wi-Fi connection logs extracted from the existing Wi-Fi infrastructure. When compared with the conventional discrete-time Markov Chain Model (MCM), the hazard-based modelling approach was able to better capture the duration dependent nature of the transition probabilities as well as incorporate and quantify the influence of the local environment on occupancy transitions. The work has demonstrated that this approach enables a convenient and flexible incorporation of urban dependenciesleading to accurate occupancy predictions whilst providing the ability to interpret the impacts of urban systems on building occupancy. Keywords: Urban system; Competing risk hazard model; Building occupancy simulation; Wi4 Fi connection data
Manca F, Sivakumar A, Daina N, et al., 2020, Modelling the influence of peers’ attitudes on choice behaviour: theory and empirical application on electric vehicle preferences, Transportation Research Part A: Policy and Practice, Vol: 140, Pages: 278-298, ISSN: 0191-2607
While the importance of social influence on transport-related choices is commonly acknowledged within the transport and travel behaviour research community, there remain several challenges in modelling influence in practice. This paper proposes a new analytical approach to measure the effects of attitudes of peers on the decision making process of the individual. Indeed, while most of the previous literature focused its attention on capturing conformity to a certain real or hypothetical choice, we investigate the subtle effect of attitudes that underlies this choice. Specifically, the suggested measure enables us to model the correlated effect that might indirectly affect the individual’s choice within a social group. It combines detailed information on the attitudes in the individual’s social network and the social proximity of the individuals in the social network. To understand its behavioural implications on the individual’s choice, the individual’s peer attitude variable is tested in different components of a hybrid choice model. Our results show that the inclusion of this variable indirectly affects the decision making process of the individual as the peers’ attitudes are significantly related to the latent attitude of the individual. On the other hand, it does not seem to directly affect the utility of an alternative as a source of systematic heterogeneity nor does it work as a manifestation of the latent variable, i.e. as an indicator.
Li T, Guo F, Krishnan R, et al., 2020, Right-of-way reallocation for mixed flow of autonomous vehicles and human driven vehicles, Transportation Research Part C: Emerging Technologies, Vol: 115, ISSN: 0968-090X
Autonomous Vehicles (AVs) are bringing challenges and opportunities to urban traffic systems. One of the crucial challenges for traffic managers and local authorities is to understand the nonlinear change in road capacity with increasing AV penetration rate, and to efficiently reallocate the Right-of-Way (RoW) for the mixed flow of AVs and Human Driven Vehicles (HDVs). Most of the existing research suggests that road capacity will significantly increase at high AV penetration rates or an all-AV scenario, when AVs are able to drive with smaller headways to the leading vehicle. However, this increase in road capacity might not be significant at a lower AV penetration rate due to the heterogeneity between AVs and HDVs. In order to investigate the impacts of mixed flow conditions (AVs and HDVs), this paper firstly proposes a theoretical model to demonstrate that road capacity can be increased with proper RoW reallocation. Secondly, four different RoW reallocation strategies are compared using a SUMO simulation to cross-validate the results in a numerical analysis. A range of scenarios with different AV penetration rates and traffic demands are used. The results show that road capacity on a two-lane road can be significantly improved with appropriate RoW reallocation strategies at low or medium AV penetration rates, compared with the do-nothing RoW strategy.
Kim I-C, Daina N, Sivakumar A, 2020, DECISION SUPPORT TOOL FOR OPTIMAL CHARGING SCHEDULING FOR INDIVIDUAL ELECTRIC VEHICLE USERS, 4th International Conference on Smart Grid and Smart Cities
Pawlak J, Imani AF, Sivakumar A, 2020, A microeconomic framework for integrated agent-based modelling of activity-travel patterns and energy consumption, Procedia Computer Science, Vol: 170, Pages: 785-790, ISSN: 1877-0509
The sophistication in the demand management approaches in both transport and energy sectors and their interaction call for modelling approaches that consider both sectors jointly. For agent-based microsimulation models of travel demand and energy consumption, this implies the necessity to ensure consistent representation of user behaviour with respect to mobility and energy consumption behaviours across the model components. Therefore this paper proposes a microeconomic framework, termed the HOT model (Home, Out-of-home, Travel) grounded in the goods-leisure paradigm, but extended to incorporate emerging activity-travel behaviour patterns and their energy consumption implications. We discuss how the model can be operationalised and embedded within agent-based frameworks with a case study using time use and energy consumption data from the UK.
Zhu L, Krishnan R, Guo F, et al., 2019, Early identification of recurrent congestion in heterogeneous urban traffic, IEEE Intelligent Transportation Systems Conference - ITSC 2019, Publisher: IEEE, Pages: 1-6
Urban traffic congestion has become a criticalissue that not only affects the quality of daily lives but alsoharms the environment and economy. Traffic patterns arerecurrent in nature, so is congestion. However, little attentionhas been paid to the development of methods that wouldenable early warning of the formation of congestion and itspropagation. This paper proposes a method for automatedearly congestion detection operating over time horizons rangingfrom half an hour to three hours. The method uses a deeplearning technique, Convolutional Neural Networks (CNN), andadapts it to the specific context of urban roads. Empiricalresults are reported from a busy traffic corridor in the city ofBath. Comprehensive evaluation metrics, including DetectionRate, False Positive Rate and Mean Time to Detection, areused to evaluate the performance of the proposed methodcompared to more conventional machine learning methodsincluding Feed-forward Neural Network and Random Forest.The results indicate that recurrent congestion can indeed bepredicted before it occurs and demonstrates that CNN basedmethod offers superior detection accuracy compared to theconventional machine learning methods in this context.
Zhu L, Krishnan R, Sivakumar A, et al., 2019, Traffic monitoring and anomaly detection based on simulation of Luxembourg Road network, IEEE Intelligent Transportation Systems Conference - ITSC 2019, Publisher: IEEE, Pages: 1-6
Traffic incidents which commonly result fromtraffic accidents, anomalous construction events and inclementweather can cause a wide range of negative impacts on urbanroad networks. Developing a high efficiency and transferabletraffic incident detection system plays an important role insolving the imbalance caused by traffic incidents betweentraffic demand and capacity. However, the existing literatureon transferability of traffic incident detection is rather limited.The objective of this paper is to provide an accurateand transferable incident detection approach based on therelationship between traffic variables and observed trafficincidents, in particular at a network level. We propose a deeplearning based method which has been calibrated using partof the collected traffic variables and the pre-assigned trafficincidents and then tested against the rest of the dataset. Theproposed method is compared to other benchmarks commonlyused in traffic incident detection, in terms of detection rate, falsepositive rate, f-measurement and detection time. The resultsindicate that the proposed method is significantly promising fortraffic incident detection with high accuracy and transferabilitycompared to the more widely used techniques in the literature.
Manca F, Sivakumar A, Polak J, 2019, Exploring the effect of social influence and live social interactions on the adoption of bike sharing in a student population, International Choice Modelling Conference 2019
Manca F, Sivakumar A, Polak J, 2019, The effect of social influence and social interactions on the adoption of a new technology: The use of bike sharing in a student population, Transportation Research Part C: Emerging Technologies, Vol: 105, Pages: 611-625, ISSN: 0968-090X
The present study investigates how social influence and social interactions can affect the adoption of new technologies, using stated preference (SP) survey data combined with an “accelerated reality” experience of social interaction among the respondents. Specifically, the intention to use a pro-environmental transport mode (the bike sharing) during a public transport strike within a cohort of students has been analysed. Previous studies have modelled social influence effects using SP data by providing a hypothetical scenario with simulated interactions or information about social conformity processes (i.e. social adoption) during the survey. In our paper, in addition to the impact of assumed social norms, the effect of live/real social interactions is included in the survey. SP survey is developed to investigate the effect of Level-of-Service attributes on the hypothetical choices in the scenario of a public transport strike. Besides the pre-defined attributes characterising the alternatives in the SP design, the survey includes techniques to acquire information on conformity and social interactions. Specifically, the interviewees undertake a before and after stated preference experiment (SP1 and SP2), with a period of group discussion in between the two parts. This SP experiment involves different cognitive and interpersonal mechanisms, such as the functional information exchange on benefits and drawbacks of cycling and bike sharing. The aim is to establish whether hypothetical scenarios of social conformity are different from real/live social interactions and whether these social influence processes actually affect the individuals' mode choice. A joint SP1/SP2 mixed logit (ML) model has been estimated to explore the choice behaviour of individuals and allows us to incorporate the inertia/propensity to change behaviour between SP1 and SP2. Moreover, considering the “Reflexive Layers of Influence” (RLI) framework, the processes generated by
Dong Y, Polak J, Sivakumar A, et al., 2019, Disaggregate short-term location prediction based on recurrent neural network and an agent-based platform, Transportation Research Record: Journal of the Transportation Research Board, Vol: 2673, Pages: 657-668, ISSN: 0361-1981
With the growing popularity of mobile and sensory devices, there has been a strong research interest in short-term disaggregate-level location prediction. Such predictive models have huge application potential in several sectors to change and improve people’s daily life and experience. Existing methods in this research stream have mainly focused on the prediction of sequence of location, with valuable temporal information overlooked. In addition, data limitations have constrained the development and understanding from different algorithms. In this paper, the authors propose a recurrent neural network-based method (RNN and LSTM, long short-term memory) for the next and future location prediction. This model predicts the sequence in time, thus it can predict both when and where an individual will be in the future and the duration of the stay at each location. The predictive model is developed based on an agent-based simulation platform that can produce realistic spatial-temporal trajectory data at the individual level. Analysis of the simulated data has shown that RNN and LSTM are capable of predicting future locations with better results than other comparative methods, especially for agents with high location variability. Online prediction with true location information fed into the model later in the day would greatly improve the predicted results. However, significant variations can be observed at the zonal level, with all methods performing much better on frequently visited locations than less visited locations or irregular visits.
Li T, Guo F, Krishnamoorthy R, et al., 2019, Right-of-Way Reallocation for Mixed Flow of Autonomous Vehicles and Human Driven Vehicles, 51st Annual Conference of the Universities-Transport-Study-Group (UTSG)
Wu C, Le Vine S, Sivakumar A, et al., 2019, Traveller preferences for free-floating carsharing vehicle allocation mechanisms, Transportation Research Part C: Emerging Technologies, Vol: 102, Pages: 1-19, ISSN: 0968-090X
Free-floating carsharing (FFCS) fleets are inherently volatile spatio-temporally, which presents both a logistical challenge for operators and a service reliability issue for customers. In this study we present a stated-choice survey to investigate the attractiveness to customers of two mechanisms for managing fleet volatility: Virtual Queuing (VQ) and Guaranteed Advance Reservation (GAR). We investigate socio-demographic features and “Big Five” personality traits that are associated ceteris paribus with choosing to use the existing FFCS service model, willingness-to-pay (WTP) for VQ and GAR, and risky-choice behaviour under the uncertainty of FFCS systems. Data (n = 289; 232 employed in analysis) are sourced from existing users of a FFCS service in London, UK. Within the survey context, we found that customers are on average not willing to pay for VQ (i.e. negative WTP), however have £0.54 per journey WTP for GAR, with low-frequency FFCS users and users scoring highly on the Big Five “Conscientiousness” dimension having larger WTP for GAR. When analysing the two dimensions of uncertainty, we found that respondents exhibit risk-seeking behaviour towards price and weaker and insignificant risk-aversion towards walking time. This pattern holds across the three standard model types of nonlinear risky-choice behaviour that we investigated. The results are intended to be useful both to policymakers and carsharing operators who are likely, as the industry matures, to seek mechanisms to differentiate their service offers to better serve individual market segments with distinctive characteristics.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.