Search

Natural disasters, such as floods, tornadoes, or wildfires, are increasingly pervasive as the Earth undergoes global warming. It is difficult to predict when and where an incident will occur, so timely emergency response is critical to saving the lives of those endangered by destructive events. Fortunately, technology can play a role in these situations. Social media posts can be used as a low-latency data source to understand the progression and aftermath of a disaster, yet parsing this data is tedious without automated methods. Prior work has mostly focused on text-based filtering, yet image and video-based filtering remains largely unexplored. In this work, we present the Incidents1M Dataset, a large-scale multi-label dataset which contains 977,088 images, with 43 incident and 49 place categories. We provide details of the dataset construction, statistics and potential biases; introduce and train a model for incident detection; and perform image-filtering experiments on millions of images on Flickr and Twitter. We also present some applications on incident analysis to encourage and enable future work in computer vision for humanitarian aid. Code, data, and models are available at http://incidentsdataset.csail.mit.edu.

This thesis report, one of a set of two reports, describes a novel way to detect incidents that could occur in the daily life of the elderly. Unlike most systems already implemented in this field, this system does not use any wearable (positioning) sensors and works off an Single Board Computer (SBC).Independent of both of these systems is a system for reassurance to alleviate distress., This is the public version of the report. The original report is shared with the jury., Electrical Engineering

Non-recurrent congestion disrupts normal traffic operations and lowers travel time (TT) reliability, which leads to many negative consequences such as difficulties in trip planning, missed appointments, loss in productivity, and driver frustration. Traffic incidents are one of the six causes of non-recurrent congestion. Early and accurate detection helps reduce incident duration, but it remains a challenge due to the limitation of current sensor technologies. In this paper, we employ a recurrence-based technique, the Quadrant Scan, to analyse time series traffic volume data for incident detection. The data is recorded by multiple sensors along a section of urban highway. The results show that the proposed method can detect incidents better by integrating data from the multiple sensors in each direction, compared to using them individually. It can also distinguish non-recurrent traffic congestion caused by incidents from recurrent congestion. The results show that the Quadrant Scan is a promising algorithm for real-time traffic incident detection with a short delay. It could also be extended to other non-recurrent congestion types.

The Indiana Department of Transportation (INDOT) has approximately 500 digital cameras along highways in populated areas of Indiana. These cameras are used to monitor traffic conditions around the clock, all year round. Currently, the videos from these cameras are observed one-by-one by human operators looking for traffic conditions and incidents. The main objective of this research was to develop an automatic, real-time system to monitor traffic conditions and detect incidents automatically. The Transportation and Autonomous Systems Institute (TASI) of the Purdue School of Engineering and Technology at Indiana University-Purdue University Indianapolis (IUPUI) and the Traffic Management Center of INDOT developed a system that monitors the traffic conditions based on the INDOT CCTV video feeds. The proposed system performs traffic flow estimation, incident detection, and classification of vehicles involved in an incident. The research team designed the system, including the hardware and software components added to the existing INDOT CCTV system; the relationship between the added system and the currently existing INDOT system; the database structure for traffic data extracted from the videos; and a user-friendly, web-based server for showing the incident locations automatically. The specific work in this project includes vehicle-detection, road boundary detection, lane detection, vehicle count over time, flow-rate detection, traffic condition detection, database development, web-based graphical user interface (GUI), and a hardware specification study. The preliminary prototype of some system components has been implemented in the Development of Automated Incident Detection System Using Existing ATMS CCT (SPR-4305).

Nowadays, with the improvement of the hardware devices and its processing capabil- ities as well as the big data field, a new promising stage in the Artificial Intelligence is born. Hence, a lot of useful applications involving AI are being developed which enable automatizing activities, patterns and voice recognition, pattern and image classification. . . Taking all this into consideration, this thesis intended to develop an incident detector and classifier using data collected by the SimRa App. This App records the rides from cyclists using the accelerometer sensor of the smartphone where it runs. First, the dataset will be overviewed to get an insight of how many useful data is available. Then a data extraction and a subsequent signal enhancement, with filtering and other signal processing techniques like ICA, will be performed in order to improve the quality of the signal that will feed the end classifier. Afterwards, a data adaptation and the developing of several classifiers will be done in order to obtain a reliable incident detector and a satisfactory incident classifier. Finally, an evaluation of the full algorithm processing time will be held as a means to know the viability of the deployment in the SimRa smartphone App.

This capstone evaluates the capabilities and potential usefulness of a Security Information and Event Management (SIEM) system in the detection of malicious network activities. The emphasis of this project was to select and configure a Free and Open Source SIEM (FOSS) to perform automated detection and alerting of malicious network events based upon predefined indicators of compromise. To test these functionalities, a virtual lab network consisting of a combination of Windows servers and Windows and Linux workstations was built to provide a proof-of-concept environment for testing the chosen FOSS SIEM. From within the lab network, a series of malicious cyber actions were executed to evaluate how well our configured FOSS solution detected and reported them. As SIEM solutions are increasingly deployed to help automate cyber defense, we hope this study motivates the adoption of FOSS solutions by organizations that may not be able to afford a commercial solution, or—perhaps— may simply prefer the advantages of free and open-source solutions., http://archive.org/details/siemenabledcyber1094560443, Outstanding Thesis, Chief Petty Officer, United States Navy, Petty Officer First Class, United States Navy, Approved for public release; distribution is unlimited.

Based on incident characteristics and quality of loop data and floating car data, a new incident detection algorithm is designed based on floating car data. This new algorithm can detect incidents on lane level by comparing the number of lane changes for a situation without an incident with a situation with a possible incident. Floating car data can give information about the number of lane changes if the accurancy is high. The floating car data is used as input for the new algorithm. The results of this new algorithm are comparable or better than the current McMaster algorithm, depending on the available penetration rate of the floating car data.

Based on incident characteristics and quality of loop data and floating car data, a new incident detection algorithm is designed based on floating car data. This new algorithm can detect incidents on lane level by comparing the number of lane changes for a situation without an incident with a situation with a possible incident. Floating car data can give information about the number of lane changes if the accurancy is high. The floating car data is used as input for the new algorithm. The results of this new algorithm are comparable or better than the current McMaster algorithm, depending on the available penetration rate of the floating car data.

The decision to select the best Intelligent Transportation System (ITS) technologies from available options has always been a challenging task. The availability of connected vehicle/automated vehicle (CV/AV) technologies in the near future is expected to add to the complexity of the ITS investment decision-making process. The goal of this research is to develop a multi-criteria decision-making analysis (MCDA) framework to support traffic agencies’ decision-making process with consideration of CV/AV technologies. The decision to select between technology alternatives is based on identified performance measures and criteria, and constraints associated with each technology. Methods inspired by the literature were developed for incident/bottleneck detection and back-of-queue (BOQ) estimation and warning based on connected vehicle (CV) technologies. The mobility benefits of incident/bottleneck detection with different technologies were assessed using microscopic simulation. The performance of technology alternatives was assessed using simulated CV and traffic detector data in a microscopic simulation environment to be used in the proposed MCDA method for the purpose of alternative selection. In addition to assessing performance measures, there are a number of constraints and risks that need to be assessed in the alternative selection process. Traditional alternative analyses based on deterministic return on investment analysis are unable to capture the risks and uncertainties associated with the investment problem. This research utilizes a combination of a stochastic return on investment and a multi-criteria decision analysis method referred to as the Analytical Hierarchy Process (AHP) to select between ITS deployment alternatives considering emerging technologies. The approach is applied to an ITS investment case study to support freeway bottleneck management. The results of this dissertation indicate that utilizing CV data for freeway segments is significantly more cos

In the current Intelligent Transportation Systems (ITS) traffic monitoring and incident detection are usually supported with mostly traditional and relatively slow reactivity technologies. In this paper we propose a new service, namely THOR (Traffic monitoring Hybrid ORiented service), able to combine two different wireless technologies and to provide real time information about vehicular traffic monitoring and incident detection. THOR relies on LTE (Long Term Evolution) and Dedicated Short Range Communication based VANETs (Vehicular ad-hoc NETworks) in a hybrid approach, which is compliant with ITS standards. This hybrid networking approach can be deployed today and can be ready for tomorrow VANET technology. We test THOR by simulations in a scenario with vehicle flows synthesized from real measured vehicular traffic traces. We provide an LTE load analysis and an assessment of incident detection capabilities. Our results are promising in terms of reactivity, precision and network traffic load sustainability.

In the current Intelligent Transportation Systems (ITS) traffic monitoring and incident detection are usually supported with mostly traditional and relatively slow reactivity technologies. In this paper we propose a new service, namely THOR (Traffic monitoring Hybrid ORiented service), able to combine two different wireless technologies and to provide real time information about vehicular traffic monitoring and incident detection. THOR relies on LTE (Long Term Evolution) and Dedicated Short Range Communication based VANETs (Vehicular ad-hoc NETworks) in a hybrid approach, which is compliant with ITS standards. This hybrid networking approach can be deployed today and can be ready for tomorrow VANET technology. We test THOR by simulations in a scenario with vehicle flows synthesized from real measured vehicular traffic traces. We provide an LTE load analysis and an assessment of incident detection capabilities. Our results are promising in terms of reactivity, precision and network traffic load sustainability.

Traffic congestion in urban areas is an increasing problem around the world. Traffic incidents (such as accidents) are considered as the major source of traffic congestion. Traffic incidents have negative impacts on traffic flow, air pollution and fuel consumption. As a result, increasing interest in finding new techniques to deal with this issue has been shown. Traffic incident-management systems can decrease the effect of such events and keep roads capacity as close as possible to normal levels. Incident detection system is an important part of any incident management system. This thesis proposes a new approach to incident detection in urban traffic networks using fuzzy logic theory with the objective of reducing traffic delays and increasing road safety. The proposed detection system can be then integrated with a traffic incident management system to reduce traffic congestion related to non-recurrent incident situations. A methodology has been established based on fuzzy logic for detecting incident status in urban areas using detector accumulative count differences. Three fuzzy models were developed and evaluated using simulated data (generated using the commercial software: PTV VISSIM by PTV Group). The fuzzy models can detect incident status on a regular basis (every minute). Performance measures were introduced to capture the capabilities of the suggested models in detecting incidents. The dissertation concludes with a summary of the major findings, recommendations and future research.

This paper presents a video-based approach to detect the presence of parked vehicles in street lanes. Potential applications include the detection of illegally and double-parked vehicles in urban scenarios and incident detection on roads. The technique extracts information from low-level feature points (Harris corners) to create spatiotemporal maps that describe what is happening in the scene. The method neither relies on background subtraction nor performs any form of object tracking. The system has been evaluated using private and public data sets and has proven to be robust against common difficulties found in closed-circuit television video, such as varying illumination, camera vibration, the presence of momentary occlusion by other vehicles, and high noise levels. © 2011 IEEE.

This paper presents a video-based approach to detect the presence of parked vehicles in street lanes. Potential applications include the detection of illegally and double-parked vehicles in urban scenarios and incident detection on roads. The technique extracts information from low-level feature points (Harris corners) to create spatiotemporal maps that describe what is happening in the scene. The method neither relies on background subtraction nor performs any form of object tracking. The system has been evaluated using private and public data sets and has proven to be robust against common difficulties found in closed-circuit television video, such as varying illumination, camera vibration, the presence of momentary occlusion by other vehicles, and high noise levels. © 2011 IEEE.

Improving the performance of a incident detection system was essential to minimize the effect of incidents. A new method of incident detection was brought forward in this paper based on an in-car terminal which consisted of GPS module, GSM module and control module as well as some optional parts such as airbag sensors, mobile phone positioning system (MPPS) module, etc. When a driver or vehicle discovered the freeway incident and initiated an alarm report the incident location information located by GPS, MPPS or both would be automatically send to a transport management center (TMC), then the TMC would confirm the accident with a closed-circuit television (CCTV) or other approaches. In this method, detection rate (DR), time to detect (TTD) and false alarm rate (FAR) were more important performance targets. Finally, some feasible means such as management mode, education mode and suitable accident confirming approaches had been put forward to improve these targets.

Traffic incidents are non-recurring events that can cause a temporary reduction in roadway capacity. They have been recognized as a major contributor to traffic congestion on our national highway systems. To alleviate their impacts on capacity, automatic incident detection (AID) has been applied as an incident management strategy to reduce the total incident duration. AID relies on an algorithm to identify the occurrence of incidents by analyzing real-time traffic data collected from surveillance detectors. Significant research has been performed to develop AID algorithms for incident detection on freeways; however, similar research on major arterial streets remains largely at the initial stage of development and testing. This dissertation research aims to identify design strategies for the deployment of an Artificial Neural Network (ANN) based AID algorithm for major arterial streets. A section of the US-1 corridor in Miami-Dade County, Florida was coded in the CORSIM microscopic simulation model to generate data for both model calibration and validation. To better capture the relationship between the traffic data and the corresponding incident status, Discrete Wavelet Transform (DWT) and data normalization were applied to the simulated data. Multiple ANN models were then developed for different detector configurations, historical data usage, and the selection of traffic flow parameters. To assess the performance of different design alternatives, the model outputs were compared based on both detection rate (DR) and false alarm rate (FAR). The results show that the best models were able to achieve a high DR of between 90% and 95%, a mean time to detect (MTTD) of 55-85 seconds, and a FAR below 4%. The results also show that a detector configuration including only the mid-block and upstream detectors performs almost as well as one that also includes a downstream detector. In addition, DWT was found to be able to improve model performance, and the use of historical dat

In this paper, the authors describe the successful combination of a low- level, vision-based surveillance system with a high-level, symbolic reasoner based on dynamic belief networks. This prototype system provides robust, high-level information about traffic scenes. The machine vision component of the system employs a correlation-based tracker and a physical motion model using a Kalman filter to extract vehicle trajectories over a sequence of traffic scene images. The symbolic reasoning component uses a dynamic belief network to make inferences about traffic events. In this paper, the authors discuss the key tasks of the vision and reasoning components as well as their integration into a working prototype.

In this paper, the authors describe the successful combination of a low- level, vision-based surveillance system with a high-level, symbolic reasoner based on dynamic belief networks. This prototype system provides robust, high-level information about traffic scenes. The machine vision component of the system employs a correlation-based tracker and a physical motion model using a Kalman filter to extract vehicle trajectories over a sequence of traffic scene images. The symbolic reasoning component uses a dynamic belief network to make inferences about traffic events. In this paper, the authors discuss the key tasks of the vision and reasoning components as well as their integration into a working prototype.

69A3551747112, The traffic patterns of your daily commute are usually predictable. Folks working on road safety, traffic operations, accessibility, or related transportation improvements can use the consistent data to implement the most effective traffic management strategies.

DTFH61-16-D00030, Traffic incident detection is a crucial task in traffic management centers (TMCs) that typically manage large highway networks with limited staff. An effective automatic incident-detection approach could benefit TMCs by helping to report abnormal events in a timely and accurate manner and optimize operating resources. During the past decades, researchers have made significant progress in developing such automatic approaches. Nevertheless, the majority of the developed approaches have shown limited success in the field, largely because of concerns about their often-costly false alarms (e.g., misdispatching response teams to a nonexistent incident). Fortunately, recent advances in artificial intelligence (AI) are expected to provide opportunities for improving conventional TMC operations. This project aimed to propose an AI-based incident-detection framework that can leverage large-scale sensor data along with advanced learning algorithms to improve the performance of incident detection. Researchers investigated the generic algorithmic problems in designing a detection approach and emphasized the architecture of the AI-based detection framework by including learning and evolving capabilities. The proposed framework was assessed with a fully controlled experiment in simulation that consisted of numerous traffic and incident scenarios. The results indicated that the proposed AI-based framework achieved higher detection rates, lower false alarm rates, and shorter time to detect the incidents in the studied scenarios than conventional approaches. Some extensions of the proposed framework are also discussed.

RR97/CB071, This report provides an update on the state-of-the-art technologies with intrusion and obstacle detection capabilities for rail rights of way (ROW) and crossings. A workshop entitled Intruder and Obstacle Detection Systems (IODS) for Railroads Requirements was held in 1998, and the Volpe National Transportation Systems Center published the proceedings in 2001. A suite of possible alternative detection technology systems were then field-tested; the results were published in 2003. A host of novel approaches to detection involving existing and emerging technologies have since appeared. This report identifies these new non-track circuit based approaches and methods of identifying obstacles and intruders on the ROW and at crossings. The results obtained from this analysis provide a technology update for the Federal Railroad Administration, as well as recommend potential technology concepts for future field testing. The application of intrusion and obstacle detection or remote sensing technologies would serve to improve the safety of rail passengers and road users, as well as protect the general population and environment from the risks associated with hazmat shipments, and aid in the relief of congestion by reducing the number of incidents and delays due to those incidents.

This research examines a range of incident detection technologies to determine a recommended combination of approaches for use in the Utah Department of Transportation (UDOT) Advanced Traffic Management System (ATMS). The technologies that were examined are computer-based Automatic Incident Detection (AID), Video Image Processing (VIP), and detection by cellular telephone call-ins.

21-8221, With the advancement of automated vehicle technologies, it is critical to understand the knowledge gap among drivers on the limitations and safety restrictions of existing advanced driving assistance systems (ADAS), which contributes to dangerous driving habits and misjudgments. For example, some ADAS include adaptive cruise control, but many drivers do not know that this feature may not function as expected in response to stationary objects. Lane departure warning systems do not always register lane markings or pavement edges with damage or covered with snow. This kind of misunderstanding has led to some highly publicized crashes. Currently, limited information is available related to crashes involving ADAS since there is no proper distinction for this kind of vehicle in current crash reporting. There is also a major concern to understand the cause and fault behind a crash involving this kind of vehicle. Without sufficient knowledge about the functionality of the technology, it is difficult for traffic incident management (TIM) personnel to determine whether the ADAS feature of a vehicle impacted a traffic incident. This study will help TIM personnel better understand ADAS technology by providing a database of commercially available vehicles incorporating this technology and training on terminology and limitations of ADAS. These findings can be used by the Utah Department of Transportation to educate both drivers and first responders.

69A3551747133, Unmanned aerial vehicles (drones) can be used in traffic and road monitoring applications. The authors investigate methods for routing a drone that is capable of simultaneously detecting road conditions and estimating traffic densities. First, the authors analyze the impact of incidents on model parameters and the difficulty in detecting capacity drops under congested conditions using speed-density observations. Then, the authors propose a framework for adaptively navigating a drone to minimize the uncertainty on parameter and traffic state estimates. This framework (1) assimilates drone observations and local sensor data, (2) quantifies the uncertainty on road network conditions and traffic states, and (3) navigates the drone towards targeted observations that minimize this uncertainty. Compared to estimation without a drone, the authors show that they can obtain significant improvement in traffic estimation and incident detection by adaptively navigating a mobile sensor that is capable of detecting capacity drops. In the future, the proposed drone navigation framework will be used for evaluating the benefit of parcel delivery systems that rely on drone detection and delivery capabilities.

Grant Award Number 2008-ST-061-TS003, "The University of Arkansas (UA) team is responsible for investigating practices of, hazardous material transportation in the private sector. The UA team is a sub‐contractor, to the project “Petrochemical Transportation Security, Development of an Interactive, Petrochemical Incident Location System (PILS), DH‐08‐ST‐061‐004” with the PI, institution being Texas Southern University National Transportation Security Center of, Excellence. This Interim Report presents synthesis of research activities in the relevant, area and overview of technologies used by J.B. Hunt Transport."

RS517/P5014, The Federal Radionavigation Plan (FRP) delineates policies and plans for radionavigation services provided by the U.S. Government to ensure efficient use of resources and full protection of national interests. Developed jointly by the U.S. Departments of Defense and Transportation, the FRP sets forth the Federal interagency approach to the implementation and operation of radionavigation systems. The FRP is updated biennially. This eighth edition describes respective areas of authority and responsibility, and provides a management structure by which the individual operating agencies will define and meet requirements in a cost-effective manner. Moreover, this edition contains the current policy on the radionavigation systems mix. The constantly changing radionavigation user profile and rapid advancements in systems technology, require that the FRP remain as dynamic as the issues it addresses. This edition of the FRP builds on the foundation laid by previous editions and further develops national plans towards providing an optimum mix of radionavigation systems for the foreseeable future.

The goal of DIVERT is to provide management strategies during periods of freeway incidents by routing diverted traffic over city surface arterials in a planned, coordinated manner. Several detailed findings of the evaluation of this project are included throughout this report.

For visual detection at night, the aircraft must display conspicuous light signals to indicate its presence and course at sufficient distance and time remaining for the pilot to avoid a collision. Considerations about the usefulness of anticollision lights must include such factors as effective light intensity, color, flashing characteristics, field of coverage, and visual detection range as well as flying speed, airplane response, and the pilot's capability to avoid a collision., Results of simulator experiments indicate that the pilot can take evasive actions at relatively high closing speeds under daylight conditions, but no data are available about collision avoidance at night.

SP009B4U, This project was focused on identifying potential areas for Maryland’s Coordinated Highway, Action Response Team (CHART) to enhance its incident management efficiency and to maximize the, resulting benefits under existing resource constraints. Using the information from CHART and the, Maryland Accident Analysis Reporting System (MAARS), this research has identified critical factors, affecting CHART’s efficiency in incident response and clearance, and produced several reliable models to, improve its performance. This research has also produced an optimal allocation model that will enable, each operational center to best deploy available patrol vehicles along its responsible highway networks, and to select the most cost-benefit fleet size under the resource constraints., CHART can also apply the set of prediction models developed in this project to estimate the, required clearance duration of a detected incident, thereby minimizing the resulting congestion within the, impact boundaries via some real-time traffic control and information strategies. Incorporating any of those, developed models into current practice will undoubtedly enhance CHART’s operational quality and, significantly increase its effectiveness in minimizing non-recurrent congestion in this region.

LFUCG-465138, This report describes the evaluation of an experimental incident detection system implemented within the Lexington/Fayette County area by the Lexington Fayette Urban County Government Department of Traffic Engineering. The incident detection system includes 5 different types of traffic monitors on 10 different locations throughout the county including interstate and limited access facilities and signalized urban arterials. The evaluation of the system was performed over a period of three months between June and August 2005. This report summarizes the implementation strategy and provides recommendations for future implementation and expansion of the system.

This research aims to provide a timely and accurate accident detection method at intersections, which is, very important for the Traffic Management System(TMS). This research uses acoustic signals to detect, accident at intersections. A system is constructed that can be operated in two modes: two-class and multiclass., The input to the system is a three-second segment of audio signal. The output of the two-class mode, is a label of “crash” or “non-crash”. In the multi-class mode of operation, the system identifies crashes as, well as several types of non-crash incidents, including normal traffic and construction sounds. The system, is composed of three main signal processing stages: feature extraction, feature reduction, and feature, classification. Five methods of feature extraction are investigated and compared; these are based on the, discrete wavelet transform, fast Fourier transform, discrete cosine transform, real cepstral transform, and, mel frequency cepstral transform. Statistical methods are used for feature optimization and classification., Three types of classifiers are investigated and compared: the nearest mean, maximum likelihood, and, nearest neighbor methods. This study focuses on the detection algorithm development. Lab testing of the, algorithm showed that the selected algorithm can detect intersection accidents with very high accuracy.

DTFH61-95-C00040, Heavily congested urban highways in the United States have spurred the development of Freeway Management Systems (FMS). Among the goals of these systems is the improvement of traffic management and the facilitation of more rapid incident detection and response. Reduction in this time, in turn, may affect the numbers of fatalities. A statistical analysis is conducted to determine the quantitative relationship between fatalities and the accident notification time on urban freeways. Using this relationship, the impact of freeway incident detection systems on fatalities is estimated. The economic benefits of fatality reduction are also derived. References, 4 tables, 18 p.

The report presents suggestions for the design of an air traffic control (ATC) incident-reporting system aimed at maximizing the amount of corrective feedback to the ATC system. The approach taken is system-oriented rather than controller-oriented. Included is a discussion of a philosophy of corrective and punitive action relative to controller involvement in an incident. Recommendations and examples of format are included for the design of incident-report forms and incident chronology and of a checklist to be used in periodic appraisal of controller performance. Emphasis is given in format design to use of systems and human function, rather than regulatory and procedural terminology. Implementation and data-analysis techniques are also discussed.

The Redding Responder Study was initiated as a component of the Redding Incident, Management Enhancement (RIME) Program. The goals of the RIME program are to leverage, technology and communications deployments for emergency communication providers in the, RIME region by evaluating agency requirements, providing migration paths and improving, incident management. The RIME region consists of 19 counties in northern California, which, cover nearly 30% of the State’s area, approximately 25% of the State’s State Highway Lane, Miles, and less than 4% of the State’s population. RIME organizations include Caltrans District, 2, Caltrans Division of Research and Innovation, Norcal EMS, California Department of, Forestry and Fire Protection, and other local and state agencies., The Redding Responder Study was sponsored by the Caltrans’ Division of Research and, Innovation. The Western Transportation Institute at Montana State University was contracted to, conduct research and development comprising the study. Research and development was, conducted to address the needs of Caltrans District 2, based in Redding. While targeted, specifically at the needs of Caltrans District 2, consideration was given to prospective needs of, other RIME agencies and other Caltrans districts, including those in urban areas. Research and, development was conducted over a two-and-one-half year time period., The premise behind the Redding Responder Study is that the collection and transmission of, digital photographs and other incident information will enhance incident management and help, to clear incidents more quickly. Secondary benefits include those associated with the, development and implementation of a systematic methodology for collecting and documenting, incidents for future analysis and training. The principal challenges include overcoming limited, communication capability in the RIME region and achieving a desired ease of use necessary to make such a system usable in the field. While off-the-shelf hardware and software products exist, to solve related problems, such products do not adequately address these principal challenges, without further integration and development., Specific situations in which the product of this study would be used include rockslides, landslides, mudslides, earthquakes, severe weather, and other events in which roadways would, be damaged or obstructed. Such events are common in District 2, particularly during the wet, months of fall, winter and spring. Use for traffic accidents and during wild land fires would also, be likely., A Project Process Model was developed to incorporate aspects of the Systems Engineering, approach, as exemplified by the “Vee” Model and the Spiral Model for Development, which is, commonly used to minimize risk in the development of complex systems. The resulting process, model is consistent with the Caltrans’ Stages of Research Deployment. Specifically, a sequence, of prototypes was developed to refine the project concept, elicit requirements and feedback, throughout the project, and evaluate technologies and techniques for use. With each “iteration,”, Caltrans was presented with the next version of the product, and feedback was elicited to, determine necessary modifications and additional requirements. Feedback was documented and, incorporated into subsequent development.

CMUIGA2012 - CMU WO 03, The effectiveness of traditional incident detection is often limited by sparse sensor coverage, and reporting incidents to emergency response systems, is labor-intensive. This research project mines tweet texts to extract incident information on both highways and arterials as an efficient and cost-effective, alternative to existing data sources. This research report presents a methodology to crawl, process and filter tweets that are accessible by, the public for free. Tweets are acquired from Twitter using the REST API in real time. The process of adaptive data acquisition establishes a, dictionary of important keywords and their combinations that can imply traffic incidents (TI). A tweet is then mapped into a high dimensional binary, vector in a feature space formed by the dictionary, and classified into either TI related or not. All the TI tweets are then geocoded to determine their, locations, and further classified into one of the five incident categories. We apply the methodology in two regions, the Pittsburgh and Philadelphia, Metropolitan Areas. Overall, mining tweets holds great potentials to complement existing traffic incident data in a very cheap way. A small sample of, tweets acquired from the Twitter API cover most of the incidents reported in the existing data set, and additional incidents can be identified through, analyzing tweets text. Twitter also provides ample additional information with a reasonable coverage on arterials. A tweet that is related to TI and, geocodable accounts for approximately 10% of all the acquired tweets. Of those geocodable TI tweets, the majority are posted by influential users, (IU), namely public Twitter accounts owned by public agencies and media, while a small number is contributed by individual users. There is more, incident information provided by Twitter on weekends than on weekdays. Within the same day, both individuals and IUs tend to report incidents more, frequently during the day time than at night, especially during traffic peak hours. Individual tweets are more likely to report incidents near the center of, a city, and the volume of information significantly decays outwards from the center. We develop a prototype web application to allow users extract, both real-time and historical incident information and visualize it on the map. The web application will be tested in PennDOT transportation, management centers., Author ORCID information: http://orcid.org/0000-0001-8716-8989

The objectives of this study are to (1) identify data sets available to MassDOT that can be used for realtime incident detection; (2) investigate how data from different sources can be integrated to improve incident detection; and (3) develop guidance for establishing trigger points to alert Highway Operations Center (HOC) operators about incidents on the road. Speed data available through the Regional Integrated Transportation Information (RITIS) platform are used for developing two alternative strategies: (a) an Artificial Intelligence (AI) model using supervised learning based on Long Short-Term Memory (LSTM) and Variational Autoencoders (VAE) layers for classifying records as normal events or incidents, and (b) an empirical rule-based method using historical speeds to establish threshold values, below which an alarm is issued requiring the HOC operator’s attention. Results on the AI model and a verified incident data set indicate a False Alarm Rate (FAR) of 0.0069% and a detection rate of 91.70%. For the empirical rule-based model, a 30-day off-line “field-test” was conducted for June 2021. Most of the events recorded by the MassDOT HOC were detected, and for most of these events the detection time was well before the “SENT-ON” time recorded in the HOC incident database.