Integrating mobile agent technology with multi-agent systems for distributed traffic detection and management systems

Agent technology is rapidly emerging as a powerful computing paradigm to cope with the complexity in dynamic distributed systems, such as traffic control and management systems. However, while a number of agent-based traffic control and management systems have been proposed and the multi-agent systems have been studied, to the best of our knowledge, the mobile agent technology has not been applied to this field. In this paper, we propose to integrate mobile agent technology with multi-agent systems to enhance the ability of the traffic management systems to deal with the uncertainty in a dynamic environment. In particular, we have developed an IEEE FIPA compliant mobile agent system called Mobile-C and designed an agent-based real-time traffic detection and management system (ABRTTDMS). The system based on Mobile-C takes advantages of both stationary agents and mobile agents. The use of mobile agents allows ABRTTDMS dynamically deploying new control algorithms and operations to respond unforeseen events and conditions. Mobility also reduces incident response time and data transmission over the network. The simulation of using mobile agents for dynamic algorithm and operation deployment demonstrates that mobile agent approach offers great flexibility in managing dynamics in complex systems.     

Validation of an agent based model using a participatory simulation gaming approach: The case of city logistics

Agent-based modeling is used for simulating the actions and interactions of autonomous entities aiming to assessing their effects on the system as a whole. At an abstract level, an agent-based model (ABM) is a representation of the many simple agents and interactions among them. The decision making of the agents is based on the rules given to them. In an ABM, the model output is the result of internal decision-making and may differ with alteration in the decision path. On the contrary, with the set of rules embedded in agents, their behavior is modeled to take a ‘certain action’ in a ‘certain situation’. It suggests that the internal decision making behavior of agents is truly responsible for the model output and thus it cannot be ignored while validating ABMs. This research article focuses on the validating agents’ behavior by evaluating decision-making processes of agents. For this purpose, we propose a validation framework based on a participatory simulation game. Using this framework we engage a human player (i.e. a domain stakeholder) to allow us to collect information about choices and validate the behavior of an individual agent. A proof-of-concept game is developed for a city logistics ABM to test the framework.     

Multi-step prediction of experienced travel times using agent-based modeling

This paper develops an agent-based modeling approach to predict multi-step ahead experienced travel times using real-time and historical spatiotemporal traffic data. At the microscopic level, each agent represents an expert in a decision-making system. Each expert predicts the travel time for each time interval according to experiences from a historical dataset. A set of agent interactions is developed to preserve agents that correspond to traffic patterns similar to the real-time measurements and replace invalid agents or agents associated with negligible weights with new agents. Consequently, the aggregation of each agent’s recommendation (predicted travel time with associated weight) provides a macroscopic level of output, namely the predicted travel time distribution. Probe vehicle data from a 95-mile freeway stretch along I-64 and I-264 are used to test different predictors. The results show that the agent-based modeling approach produces the least prediction error compared to other state-of-the-practice and state-of-the-art methods (instantaneous travel time, historical average and k-nearest neighbor), and maintains less than a 9% prediction error for trip departures up to 60 min into the future for a two-hour trip. Moreover, the confidence boundaries of the predicted travel times demonstrate that the proposed approach also provides high accuracy in predicting travel time confidence intervals. Finally, the proposed approach does not require offline training thus making it easily transferable to other locations and the fast algorithm computation allows the proposed approach to be implemented in real-time applications in Traffic Management Centers.     

Studying the benefits of carpooling in an urban area using automatic vehicle identification data

Carpooling has been considered a solution for alleviating traffic congestion and reducing air pollution in cities. However, the quantification of the benefits of large-scale carpooling in urban areas remains a challenge due to insufficient travel trajectory data. In this study, a trajectory reconstruction method is proposed to capture vehicle trajectories based on citywide license plate recognition (LPR) data. Then, the prospects of large-scale carpooling in an urban area under two scenarios, namely, all vehicle travel demands under real-time carpooling condition and commuter vehicle travel demands under long-term carpooling condition, are evaluated by solving an integer programming model based on an updated longest common subsequence (LCS) algorithm. A maximum weight non-bipartite matching algorithm is introduced to find the optimal solution for the proposed model. Finally, road network trip volume reduction and travel speed improvement are estimated to measure the traffic benefits attributed to carpooling. This study is applied to a dataset that contains millions of LPR data recorded in Langfang, China for 1 week. Results demonstrate that under the real-time carpooling condition, the total trip volumes for different carpooling comfort levels decrease by 32–49%, and the peak-hour travel speeds on most road segments increase by 5–40%. The long-term carpooling relationship among commuter vehicles can reduce commuter trips by an average of 30% and 24% in the morning and evening peak hours, respectively, during workdays. This study shows the application potential and promotes the development of this vehicle travel mode.     

Can agents measure up? A comparative study of an agent-based and on-line optimization approach for a drayage problem with uncertainty

Experiments studying the behavior of agent-based methods over varying levels of uncertainty in comparison to traditional optimization methods are generally absent from the literature. In this paper we apply two structurally distinct solution approaches, an on-line optimization and an agent-based approach, to a drayage problem with time windows under two types of uncertainty. Both solution approaches are able to respond to dynamic events. The on-line optimization approach utilizes a mixed integer program to obtain a feasible route at 30-s intervals. The second solution approach deploys agents that engage in auctions to satisfy their own objectives based on the information they perceive and maintain locally. Our results reveal that the agent-based system can outperform the on-line optimization when service time duration is highly uncertain. The on-line optimization approach, on the other hand, performs competitively with the agent-based system under conditions of job-arrival uncertainty. When both moderate service time and job-arrival uncertainties are combined, the agent system outperforms the on-line optimization; however, in the case of extremely high combined uncertainty, the on-line optimization outperforms the agent-based approach.     

Multiagent architectures for intelligent traffic management systems

This paper reports our experiences with agent-based architectures for intelligent traffic management systems. We describe and compare integrated TRYS and TRYS autonomous agents, two multiagent systems that perform decision support for real-time traffic management in the urban motorway network around Barcelona. Both systems draw upon traffic management agents that use similar knowledge-based reasoning techniques in order to deal with local traffic problems. Still, the former achieves agent coordination based on a traditional centralized mechanism, while in the latter coordination emerges upon the lateral interaction of autonomous traffic management agents. We evaluate the potentials and drawbacks of both multiagent architectures for the domain, and develop some conclusions respecting the general applicability of multiagent architectures for intelligent traffic management.     

Applying a structured simulation-based methodology to assess carpooling time–space potential

Abstract

Increasing urban traffic congestion calls for the study of alternative measures. One such measure is carpooling, a system in which a person shares his private vehicle with one or more people in a commuter trip. In principle, this system could lead to potentially significant reductions in the use of private vehicles; however, in practice it has achieved limited success. In this paper, we apply a simulation-based methodology that uses aggregated data from commuter trips in an urban area to create compatible and feasible random trips. These are then analyzed through a heuristic process recursively to find grouping possibilities, thus producing indicators of carpooling potential such as the percentage of matched trips. Using this methodology, simulations are run for the Lisbon Metropolitan Area (Portugal) and results show that an increase in the number of participants in a carpooling scheme will only increase the probability of matching up to a certain point, and that this probability varies significantly with time–space trip attributes.     

A dynamic cordon pricing scheme combining the Macroscopic Fundamental Diagram and an agent-based traffic model

Pricing is considered an effective management policy to reduce traffic congestion in transportation networks. In this paper we combine a macroscopic model of traffic congestion in urban networks with an agent-based simulator to study congestion pricing schemes. The macroscopic model, which has been tested with real data in previous studies, represents an accurate and robust approach to model the dynamics of congestion. The agent-based simulator can reproduce the complexity of travel behavior in terms of travelers’ choices and heterogeneity. This integrated approach is superior to traditional pricing schemes. On one hand, traffic simulators (including car-following, lane-changing and route choice models) consider travel behavior, i.e. departure time choice, inelastic to the level of congestion. On the other hand, most congestion pricing models utilize supply models insensitive to demand fluctuations and non-stationary conditions. This is not consistent with the physics of traffic and the dynamics of congestion. Furthermore, works that integrate the above features in pricing models are assuming deterministic and homogeneous population characteristics. In this paper, we first demonstrate by case studies in Zurich urban road network, that the output of a agent-based simulator is consistent with the physics of traffic flow dynamics, as defined by a Macroscopic Fundamental Diagram (MFD). We then develop and apply a dynamic cordon-based congestion pricing scheme, in which tolls are controlled by an MFD. And we investigate the effectiveness of the proposed pricing scheme. Results show that by applying such a congestion pricing, (i) the savings of travel time at both aggregated and disaggregated level outweigh the costs of tolling, (ii) the congestion inside the cordon area is eased while no extra congestion is generated in the neighbor area outside the cordon, (iii) tolling has stronger impact on leisure-related activities than on work-related activities, as fewer agents who perform work-related activities changed their time plans. Future work can apply the same methodology to other network-based pricing schemes, such as area-based or distance-traveled-based pricing. Equity issues can be investigated more carefully, if provided with data such as income of agents. Value-of-time-dependent pricing schemes then can also be determined.     

Seattle smart traveler: dynamic ridematching on the World Wide Web

Seattle Smart Traveler (SST) is an application of World Wide Web (WWW) technology to test the concept of automated dynamic rideshare matching. In SST, the rideshare clientele interact with the rideshare system using only WWW pages. SST collects spatial and temporal trip information using a series of WWW pages, performs a match using structured query language (SQL) specifications to a database engine, and supports both the standard phone-based contact methodology as well as two new, unique e-mail-based contact methodologies. SST was operated in parallel to a traditional, regional rideshare system for one year, and the two systems were marketed to the user community on a side-by-side basis. SST and the traditional system acquired approximately the same number of new users over a nine-month test period; however, there was little overlap in the population using the two parallel systems. SST demonstrates that there is a user population that can be reached using Internet technologies for immediate/dynamic ridematching that is not reached by traditional ridematch programs. This paper also reports a new statistical model for quantifying rideshare matching and carpooling. The model is validated using the SST experimental results, and the model demonstrates that the carpooling process is a quadratic function of the number of users participating.     

Agent-based network transmission model using the properties of macroscopic fundamental diagram

The field of research that has recently come to the fore is the perimeter control, which aims to control traffic demand for a large urban area prior to controlling internal flow inside the area. Such control concept needs to be tested by simulations, hence, it is necessary to develop a model that can appropriately estimate the network-wide flow dynamics. In this paper, agent-based network transmission model (ANTM) is proposed for describing the aggregated flow dynamics over an urban area of multiple large-scale networks. The proposed model is the combination of the cell transmission model (CTM), macroscopic fundamental diagram (MFD), and agent concept. The CTM-based simulation is adopted for the simplicity considering the computation requirements for real-time feasibility. The MFD concept is applied for representing the network properties, and a new approach is taken particularly for estimating network outflow affected by both demand patterns and boundary capacity. The agent concept is applied for representing drivers’ travel behaviors. The model is compared with microscopic simulations and shows reasonable accuracy for large areas. In addition, various travel direction choice behaviors are applicable to this model. Various perimeter control policies are applicable as well, thus, the proposed model can be a useful tool for testing various control methods, in terms of reducing the congestion in urban areas.     

What encourages people to carpool? An evaluation of factors with meta-analysis

Non-household carpools (where two or more commuters from different residences travel together in the same private vehicle) bring public benefits. To encourage and incentivise it, transport practitioners and researchers must understand its private motivations and deterrents. Existing studies often report conflicting results or non-generalisable findings. Thus, a quantitative systematic review of the literature body is needed. Using meta-analysis, this study synthesised 22 existing empirical studies (representing over 79,000 observations) to produce an integrated review of the carpooling literature. The meta-analysis determined 24 non-household carpooling factors, and their effect sizes. Factors such as number of employees (\(\bar{r} = 0.42\)), partner matching programs (\(\bar{r} = 0.42\)), female (\(\bar{r} = 0.22\)) and fixed work schedule (\(\bar{r} = 0.15\)) were found to have strong effects on carpooling while judgmental factors (such as the motivation to save costs) only exhibited small influence (\(\bar{r} < 0.1\)). Based on the significant effects, the paper discussed prospects for improving carpooling uptake by developing: (i) target demographics, (ii) selling points for marketing, (iii) carpooling partner programs and (iv) multiple employer ‘super-pools’. The results warrant caution due to the small amount of studies synthesised. Transport practitioners might plan carpooling policies based on the findings; and transportation researchers might use the list of factors to model carpooling behaviour.     

Modelling commuting mode choice with explicit consideration of carpool in the choice set formation

This article investigates the carpool mode choice option in the context of overall commuting mode choice preferences. The article uses a hybrid discrete choice modelling technique to jointly model the consideration of carpooling in the choice set formation as well as commuting mode choice together with the response bias corrections through the accommodation of measurement equations. A cross-nested error structure for the econometric formulation is used to capture correlations among various commuting modes and carpool consideration in the choice set. Empirical models are estimated using a data set collected through a week-long commuter survey in Edmonton, Alberta. The empirical model reveals many behavioural details of commuting mode choice and carpooling. Interestingly, it reveals that interactions between various Travel Demand Management (TDM) tools with the carpooling option can be different at different level of decision making (choice set formation level and final choice making level).     

Optimal starting location of an HOV lane for a linear monocentric urban area

This paper analyzes the optimal starting location of a high-occupancy-vehicle (HOV) lane for a linear monocentric urban area. Both travel time and carpooling costs are taken into account. The research proposes an analytical framework for the case with a continuum demand distribution along a highway corridor. The objective is assumed to maximize social welfare of the transportation system, which is the difference between the total user benefit and travel cost. Numerical analysis via simulation experiments was conducted to seek the existence of an optimal solution. Based on the results of a sensitivity analysis, we find a specific relationship between the carpooling cost and the optimal design of the starting point of an HOV lane.     

Getting by with a little help from my friends…and family: immigrants and carpooling

While much of the scholarly literature on immigrants’ travel focuses on transit use, the newest arrivals to the United States make over twelve times as many trips by carpool as by transit. Using the 2001 National Household Travel Survey and multinomial logit mode choice models, we examine the determinants of carpooling. In particular, we focus on the likelihood of carpooling among immigrants—carpooling both within and across households. After controlling for relevant determinants of carpooling, we find that immigrants are far more likely to form household carpools than native-born adults and also are more likely than the native-born to form external carpools (outside the household). Moreover, when faced with the options of carpooling and public transit, immigrants—even recent arrivals—appear to prefer carpools over transit more strongly than the native born.     

Incorporating time and income constraints in dynamic agent-based models of activity generation and time use: Approach and illustration

Existing theories and models in economics and transportation treat households’ decisions regarding allocation of time and income to activities as a resource-allocation optimization problem. This stands in contrast with the dynamic nature of day-by-day activity-travel choices. Therefore, in the present paper we propose a different approach to model activity generation and allocation decisions of individuals and households that acknowledges the dynamic nature of the behavior. A dynamic representation of time and money allocation decisions is necessary to properly understand the impact of new technologies on day to day variations in travel and activity patterns and on performance of transportation systems. We propose an agent-based model where agents, rather than acting on the basis of a resource allocation solution for a given time period, make resource allocation decisions on a day-by-day basis taking into account day-varying conditions and at the same time respecting available budgets over a longer time horizon. Agents that share a household interact and allocate household tasks and budgets among each other. We introduce the agent-based model and formally discuss the properties of the model. The approach is illustrated on the basis of simulation of behavior in time and space.     

A study of carpooling behaviour using a stated preference web survey in selected cities of India

This paper analyses how people perceive the idea of carpooling and evaluate preferences while making a decision to join a carpool. Analysing data from a web-based stated preference survey in India reveals that cognitive attitudes play a significant role in evaluating the perceived advantages and disadvantages of carpooling whereas intentions to carpool are associated with perceived negative evaluations. A factor analysis identifies two latent attitudinal factors: a ‘time–convenience’ factor that discourages carpooling and a ‘private–public cost’ factor that encourages carpooling. The study analyses the influential attributes – extra travel time, walking time to reach meeting point, waiting time at pickup point and cost savings – as explanatory variables for the utility of carpooling. Cost savings prove to be the most significant attribute when combined with other attributes, followed by extra travel time. The study provides the implications to policy-makers of designing promotional tools to improve the propensity of carpooling among single occupant vehicle drivers.     

Willingness to accept longer commutes for better salaries: Understanding the differences within and between couples

This paper reports on an analysis aiming to understand differences across individual people in their willingness to accept increased commuting time in return for higher salary, using Hierarchical Bayes (HB) analysis of a dataset collected in Sweden. We find that socio-demographic and attitudinal differences are significant in explaining the variations in values of time for individuals, in particular income, who drives when carpooling and hours worked per week. Additionally we also examine the values of individuals when their choices also impact on the salary and commute of their partner, finding that incomes, income differentials, driving behaviour when carpooling, division of housework and car user decisions significantly explain the values assigned to others and variations in an individual’s own values once their partner is affected. The overall richness of the results reflect the benefits that posterior analysis can bring, and highlight the computational efficiency of Bayesian methods in producing such conditionals at an individual level.     

Online distributed cooperative model predictive control of energy-saving trajectory planning for multiple high-speed train movements

The cooperative energy-efficient trajectory planning for multiple high-speed train movements is considered in this paper. We model all the high-speed trains as the agents that can communicate with others and propose a local trajectory planning control model using the Model Predictive Control (MPC) theory. After that we design an online distributed cooperative optimization algorithm for multiple train trajectories planning, under which each train agent can regulate the trajectory planning procedure to save energy using redundancy trip time through tuning ACO’s heuristic information parameter. Compared to the existing literature, the vital distinctions of our work lies not only on the online cooperative trajectory planning but also on the distributed mechanism for multiple high-speed trains. Experimental studies are given to illustrate the effectiveness of the proposed methods with the practical operational data of Wuhan-Guangzhou High-speed Railway in China.     

Carpooling and carpool clubs: Clarifying concepts and assessing value enhancement possibilities through a Stated Preference web survey in Lisbon, Portugal

The increase of urban traffic congestion calls for studying alternative measures for mobility management, and one of these measures is carpooling. In theory, these systems could lead to great reductions in the use of private vehicles; however, in practice they have obtained limited success for two main reasons: the psychological barriers associated with riding with strangers and poor schedule flexibility. To overcome some of the limitations of the traditional schemes, we proposed studying a carpooling club model with two main new features: establishing a base trust level for carpoolers to find compatible matches for traditional groups and at the same time allowing to search for a ride in an alternative group when the pool member has a trip schedule different from the usual one. A web-based survey was developed for the Lisbon Metropolitan Region (Portugal), including a Stated Preference experiment, to test the concept and confirm previous knowledge on these systems’ determinants. It was found through a binary logit Discrete Choice Model calibration that carpooling is still attached with lower income strata and that saving money is still an important reason for participating in it. The club itself does not show promise introducing more flexibility in these systems; however, it should provide a way for persons to interact and trust each other at least to the level of working colleagues.     

The GRAAL of carpooling: GReen And sociAL optimization from crowd-sourced data

Carpooling, i.e. the sharing of vehicles to reach common destinations, is often performed to reduce costs and pollution. Recent work on carpooling takes into account, besides mobility matches, also social aspects and, more generally, non-monetary incentives. In line with this, we present GRAAL, a data-driven methodology for GReen And sociAL carpooling. GRAAL optimizes a carpooling system not only by minimizing the number of cars needed at the city level, but also by maximizing the enjoyability of people sharing a trip. We introduce a measure of enjoyability based on people’s interests, social links, and tendency to connect to people with similar or dissimilar interests. GRAAL computes the enjoyability within a set of users from crowd-sourced data, and then uses it on real world datasets to optimize a weighted linear combination of number of cars and enjoyability. To tune this weight, and to investigate the users’ interest on the social aspects of carpooling, we conducted an online survey on potential carpooling users. We present the results of applying GRAAL on real world crowd-sourced data from the cities of Rome and San Francisco. Computational results are presented from both the city and the user perspective. Using the crowd-sourced weight, GRAAL is able to significantly reduce the number of cars needed, while keeping a high level of enjoyability on the tested data-set. From the user perspective, we show how the entire per-car distribution of enjoyability is increased with respect to the baselines.