Predicting delay reductions from freeway traffic diversion for incident management

Traffic incidents are a major source of traffic congestion on freeways. Freeway traffic diversion using pre-planned alternate routes has been used as a strategy to reduce traffic delays due to major traffic incidents. However, it is not always beneficial to divert traffic when an incident occurs. Route diversion may adversely impact traffic on the alternate routes and may not result in an overall benefit. This dissertation research attempts to apply Artificial Neural Network (ANN) and Support Vector Regression (SVR) techniques to predict the percent of delay reduction from route diversion to help determine whether traffic should be diverted under given conditions. The DYNASMART-P mesoscopic traffic simulation model was applied to generate simulated data that were used to develop the ANN and SVR models. A sample network that comes with the DYNASMART-P package was used as the base simulation network. A combination of different levels of incident duration, capacity lost, percent of drivers diverted, VMS (variable message sign) messaging duration, and network congestion was simulated to represent different incident scenarios. The resulting percent of delay reduction, average speed, and queue length from each scenario were extracted from the simulation output. The ANN and SVR models were then calibrated for percent of delay reduction as a function of all of the simulated input and output variables. The results show that both the calibrated ANN and SVR models, when applied to the same location used to generate the calibration data, were able to predict delay reduction with a relatively high accuracy in terms of mean square error (MSE) and regression correlation. It was also found that the performance of the ANN model was superior to that of the SVR model. Likewise, when the models were applied to a new location, only the ANN model could produce comparatively good delay reduction predictions under high network congestion level.

A model of mistaken alibis: How innocent alibi providers generate inaccurate alibis

This dissertation comprised two experiments, which addressed three main goals: (a) to test a new paradigm for measuring objectively the accuracy of alibis, (b) to explore the effectiveness of three retrieval cues (time only, location only, and time-and-location) in an alibi context, and (c) to explore the metacognitive strategies of innocent alibi providers who experience different financial incentives as well as different motivations for reporting (be informative vs. be convincing). ^ The novel paradigm appears promising: by surreptitiously controlling the whereabouts of future alibi providers during a critical time, objective accuracy measurements were in fact possible. Such accuracy measurements revealed that time-cued retrieval can be devastating to innocent alibi providers. Participants who attempted to recall their whereabouts via a time cue were significantly less accurate than participants who attempted recall via a location cue (Experiment 1). ^ Innocent alibi providers, when cued effectively, may not, however, report their memories differently from memory reporters in non-alibi contexts. When cued effectively, participants who experienced a goal of being convincing did not differ in accuracy from participants who experienced a goal of merely being informative (Experiment 2). Similarly, participants did not differ from one another in accuracy across different levels of financial incentive (Experiment 2). ^ Despite the indistinguishable accuracy rates of alibi providers and non-alibi memory reporters when retrieval was cued effectively, proffering mistaken alibis presents a real risk for innocent suspects. Future research needs to address methods by which that risk can be reduced. ^

The Influence of Psychological Predictors and Cognitive Behavioral Stress Management Intervention on Antiretroviral Therapy (ART) Adherence Among HIV-Positive Female Haitian Alcohol Users

Purpose: Over half the HIV-infected persons in the Caribbean, the second most HIV-impacted region in the world, live in Haiti. Using secondary data from a parent study, this research assessed the effects of psychological and social factors on antiretroviral therapy (ART) adherence among Haitian, HIV-positive, female alcohol users. Theoretical Foundation and Research Questions: Using the Theory of Planned Behavior/Reasoned Action and the Information, Motivation, Behavior skills model as guiding theoretical frameworks, the study examined the effectiveness of an adapted cognitive behavioral stress management (CBSM-A) intervention in improving ART adherence. The effect of psychological factors (depression, anxiety, beliefs about medicine, and social support), social factors (stigma, relationship status, and educational attainment), and alcohol on adherence to ART was assessed. Methods: The sample consisted of 116 female ART patients who were randomly assigned to the CBSM-A intervention or the wait-list control group. Participants completed intervention sessions as well as pre- and post-test assessments. Analyses of variance, t-tests, and point biserial correlations were used to test hypotheses. Results: Surprisingly, ART adherence rates significantly decreased for both groups combined [F (1, 108) = 8.79, p = .004]; there was no significant difference between the intervention and control groups with regard to the magnitude of change between baseline and post assessment. On average, depression decreased significantly among participants in the CBSM-A group only [(t (62) = 5.54, p < .001)]. For both groups combined, alcohol use significantly decreased between baseline and post-assessment [(F (1, 78) = 34.70, p < .001)]; there was no significant difference between the intervention and control groups with regard to the magnitude of change between baseline and post-assessment. None of the variables were significantly correlated with ART adherence. Discussion: Adherence to ART did not improve in this sample, nor were any of the variables significantly associated with adherence. The findings suggest that additional supportive and psychological services may be needed in order to promote higher adherence to ART among HIV-positive females. More research may be needed on this sample; a focus on mental health issues, partner conflict, family and sexual history may allow for better targeting and more successful interventions.

Predicting delay reductions from freeway traffic diversion for incident management

Traffic incidents are a major source of traffic congestion on freeways. Freeway traffic diversion using pre-planned alternate routes has been used as a strategy to reduce traffic delays due to major traffic incidents. However, it is not always beneficial to divert traffic when an incident occurs. Route diversion may adversely impact traffic on the alternate routes and may not result in an overall benefit. This dissertation research attempts to apply Artificial Neural Network (ANN) and Support Vector Regression (SVR) techniques to predict the percent of delay reduction from route diversion to help determine whether traffic should be diverted under given conditions. The DYNASMART-P mesoscopic traffic simulation model was applied to generate simulated data that were used to develop the ANN and SVR models. A sample network that comes with the DYNASMART-P package was used as the base simulation network. A combination of different levels of incident duration, capacity lost, percent of drivers diverted, VMS (variable message sign) messaging duration, and network congestion was simulated to represent different incident scenarios. The resulting percent of delay reduction, average speed, and queue length from each scenario were extracted from the simulation output. The ANN and SVR models were then calibrated for percent of delay reduction as a function of all of the simulated input and output variables. The results show that both the calibrated ANN and SVR models, when applied to the same location used to generate the calibration data, were able to predict delay reduction with a relatively high accuracy in terms of mean square error (MSE) and regression correlation. It was also found that the performance of the ANN model was superior to that of the SVR model. Likewise, when the models were applied to a new location, only the ANN model could produce comparatively good delay reduction predictions under high network congestion level.