1,720,989 research outputs found
A methodology for traffic state estimation and signal control utilizing high wireless device penetration
No full textThis paper presents a methodology for fusing data from multiple sensors, including wireless devices, to make an estimation of the state of an urban traffic network. An extended Kalman filter is employed along with a state evolution model to make estimates of the state in a discretized network. Results are presented from simulation tests of signal controllers on a network with three signalized junctions. Two signal control methods are tested: SCOOT and a machine learning junction control algorithm that employs the discretized state structure described in this paper. These tests represent lower and upper performance benchmarks and present a significant difference. The tests also demonstrate a framework for the future evaluation of the proposed methodology
An automated signalized junction controller that learns strategies by temporal difference reinforcement learning
Full text linkThis paper shows how temporal difference learning can be used to build a signalized junction controller that will learn its own strategies through experience. Simulation tests detailed here show that the learned strategies can have high performance. This work builds upon previous work where a neural network based junction controller that can learn strategies from a human expert was developed. In the simulations presented, vehicles are assumed to be broadcasting their position over WiFi giving the junction controller rich information. The vehicle’s position data are pre-processed to describe a simplified state. The state-space is classified into regions associated with junction control decisions using a neural network. This classification is the strategy and is parameterized by the weights of the neural network. The weights can be learned either through supervised learning with a human trainer or reinforcement learning by temporal difference (TD).Tests on a model of an isolated T junction show an average delay of 14.12s and 14.36s respectively for the human trained and TD trained networks. Tests on a model of a pair of closely spaced junctions show 17.44s and 20.82s respectively. Both methods of training produced strategies that were approximately equivalent in their equitable treatment of vehicles, defined here as the variance over the journey time distributions
Understanding the explanatory factors leading to variability in charity collection bank yields: implications for bank placement and logistics strategy
No full textOxfam runs a network of approximately 570 textile donation banks across the UK for second-hand textiles and books to raise revenue for good causes around the world. These banks are placed in a variety of public places, such as car parks (public and supermarket) and recycling centres. Of real interest is the impact the underlying population characteristics have on the yield volume and stock quality at the site, and whether a better understanding of this relationship can be gained using historic fill levels over time, and quality audits of the stock donated. If a relationship exists and different postcodes can be shown to yield significantly different weights and quality of stock then banks can be targeted to certain areas, and the logistics optimised accordingly.Using a database of Monthly (April 2010 – March 2011) collection records (weights of donations collected per bank) for around 500 textile banks, this paper investigates the relative impacts of certain factors (season, location, proximity to services and levels of affluence) in influencing the variability in observed banks weights across Oxfam’s charity banks in England. It also presents a model which can be used to predict the weights of donations that should be generated by a bank given the profile of the region. For each factor, a weighting coefficient was calculated to generate predicted weights for each bank.Data for 2009-2010 have also been used for analysis to understand seasonal and longer term trends. Data regarding the key factors used within the model have been derived from a range of sources (eg. Office for National Statistics). In order to identify whether areas with different demographic, geographical and economic characteristics, yield significantly different volumes of saleable stock, a range of donation stock quality audits are currently being undertaken at a range of locations around the country based on the outcomes of the analysis
Gauging HWRC performance from vehicle weigh-ticket data
No full textThis paper describes a modelling approach designed to investigate the variability in nett amenity bin weights produced by nine household waste recycling centres (HWRCs) in West Sussex, UK over a 12-month period. Compaction technique, vehicle type, site design and month were identified as key factors explaining 76% of the variability in the data. For each significant factor, a weighting coefficient was calculated to generate a predicted nett weight for every bin transaction. Analysis of predicted and observed mean bin weights suggested that three sites had similar characteristics but returned significantly different mean nett bin weights. Subsequent waste and site audits determined the possible sources of the remaining variability. Significant differences were identified in the proportions of bagged waste and dry recyclables deposited in the amenity waste stream at the sites, with significantly less observed at one site. Operational and managerial techniques (e.g. material separation, compaction frequency and site management ethos) were also identified as factors impacting on mean bin weights and general site performance. The model can be used to identify sites producing significantly different bin weights, enabling detailed ‘back-end’ waste analyses to be efficiently targeted and best practice in HWRC operation identifie
Collection and use of environmental data for transport management: a view from local authorities
No full textCollection and use of environmental data for transport management: a view from local authorities
No full textResearch undertaken by the Transportation Research Group at the University of Southampton is discussed, which studies the views, needs and requirements of local authorities as users of sensor grids for environmental monitoring. The study was undertaken through a combination of literature review and workshop activity. Representatives from local authorities were invited to participate in a workshop discussion. It was agreed that more detailed and comprehensive environmental data will be useful for transport planning, network management and traveller information provision. The detailed data can be used to provide a sound understanding of local environmental situations and to identify specific problem areas and times. For local authorities, monitoring noise level and air pollutants which have direct impact on human health such as PM
10, PM
2.5 and NO
x is more important than monitoring green house emission. Historic data on air pollution will be essential for local transport plans focusing on reduction of road-transport-related pollution. Real time environmental data will be needed to be integrated with existing traffic control and traveller information systems. Information on areas surrounding schools and hospitals are most desirable. However, there was a concern regarding negative information disseminated to the public.</p
Urban traffic state estimation for signal control using mixed data sources and the extended Kalman filter
No full textThis paper describes a methodology for fusing data from multiple sensors, including wireless devices and inductive loops, to make an estimation of the instantaneous state of an urban traffic network. An extended Kalman filter is employed along with a state evolution model to make estimates of the state in a discretized network. The instantaneous state is an estimate of the current distribution of vehicles in the network and their instantaneous speeds. Microsimulation tests were used to evaluate the performance of the state estimation on a small urban networks. These results indicate low error between the estimated state and the known ground truth
Remote automatic incident detection using inductive loops
Full text linkThis paper describes the remote automatic incident detection algorithm designed to detect abnormal periods of traffic congestion existing over single inductive loop detectors (typically 2 3 1.5 m). This algorithm identifies those detectors which show a critical increase in average loop-occupancy time per vehicle coinciding with a critical decrease in average time-gap between vehicles according to a set of rules previously defined by the operator. The rules define the maximum and minimum values of loop occupancy and time gap respectively for each detector, which when exceeded for a given duration, trigger a report of a potential traffic flow ‘abnormality’ for that time of day at that particular location on the network. Initial rules are developed by studying the 85th percentile values of loop occupancy returned by the urban traffic control system every 30 s. A real-time trial took place between 07:00 and 19:00 over 167 consecutive days involving 74 detectors situated along two sections of the A33 Bassett Avenue and A35 Winchester Road in Southampton. Over this period, 181and 334 triggers were recorded on the A33 and A35, respectively. An independent operator log showed that over the same period, 32 incidents were recorded on the A33 and 49 on the A35. The remote automatic incident detection system detected 69% and 92% of the verified incidents on the A33 and A35, respectively; the low detection rate on the A33 being mainly due to five incidents which occurred during off-peak periods causing no congestion and were therefore not detected
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
