Jaramillo-Giraldo, Erika C.

Profile Picture

Filters

2009 - 200912010 - 20141
Reset filters

Settings

Citations

Publication Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Publication
    A methodology to incorporate highway safety into transportation planning using generalized linear mixed models
    (2014) Jaramillo-Giraldo, Erika C.; Valdés-Díaz, Didier M.; College of Engineering; Figueroa Medina, Alberto; Cruzado Vélez, Ivette; Torres Saavedra, Pedro; Macchiavelli, Raúl; Department of Civil Engineering; Lorenzo, Edgardo
    Highway Safety has been identified as a significant problem worldwide. Crashes have been found as the second cause of death in the world according to the World Health Organization (2010). In fact, road crashes cost billions of dollars per year in the US alone. The US Department of Transportation has established highway safety as one of their main priorities in their Action Plan that mainly consists of establishing countermeasures and engineering strategies for the reduction of crashes. Several efforts are underway but most of the implemented strategies in many states and Puerto Rico have a reactive or a short-term planning approach. Such approaches have generated some improvements to the current system (Lovegrone, 2006). However, a proactive approach is necessary. This approach would require incorporating highway safety aspects in the decision making process from the beginning when planning alternatives are generated and crash data is unknown (de Leur, 2001). Currently, Safety Performance Functions (SPFs) are considered by many as the main tool in estimating a road’s safety and an integral part of decision making. SPFs are mathematical models that are statistically developed to conduct crash data analysis. The models attempt to explain crash occurrence on various road facilities types as a function of the traffic and geometric characteristics of these facilities. SPFs are not just valuable to the success of the reactive approach to dealing with road safety problems; they are of vital importance to the success of the proactive approach. The primary objective of the proactive approach is to ensure that road safety is an explicit priority in transportation planning policies. Several crash prediction models have been developed for site or project analysis, but very few of them are for planning purposes. Strategic, mid and short-term planning models with a wide prediction range, due to a wide range of Average Annual Daily Traffic (AADT) measures, were not found in the researched literature. A model of this type could be used to calculate the number of average crashes per type of road while considering conceptual design aspects or design changes. They could also be used in the implementation of safety devices for the whole region. These types of models were developed in this research project. This research project utilized Generalized Linear Mixed Models (GLMM) so as to use them in the incorporation of highway safety into the strategic planning process. These models can be used to forecast the rate of crashes for different planning and conceptual design scenarios. These GLMM have several advantages, in terms of predicting crash rates, including the incorporation of not only a set of known explanatory variables, but also of random effects present in the system. As a result, the model explains possible temporal correlation and spatial effects in the data. Therefore, these types of models offer great versatility in the modeling of crash rates and its related factors. The research approach included the filtering of a crash database according to a set of identified explanatory variables, estimation of parameters in a set of candidate GLMM’s, evaluation of the estimated models using several statistical tests and goodness of fit methods, and the selection of models that represent a better fit for the phenomena under study. Once the models were obtained, a methodology for their incorporation into the strategic planning process was developed and reported. Therefore, the deliverables included are GLMM crash prediction models for municipalities and different types of segments on expressways-freeways and arterials (population average and specific subject models), along with a methodology for the incorporation of these models in safety analysis which is a vital part of the strategic planning process.
  • Thumbnail Image
    Publication
    Desarrollo de un marco de referencia para abordar el tema de la seguridad en la industria de la construcción desde una perspectiva ética
    (2009) Jaramillo-Giraldo, Erika C.; Maldonado-Fortunet, Francisco; College of Engineering; González Quevedo, Antonio A.; Valdés Díaz, Didier M.; Ferrer, Jorge J.; Department of Civil Engineering; O’Neill, Efrain
    Las tasas de accidentes y muertes dentro de la industria de la construcción han sido y siguen siendo desalentadoras. En muchas ocasiones este asunto ha sido estudiado y se ha tratado de mitigar a través de aplicaciones de carácter técnico. Otras veces se han establecido múltiples reglamentaciones y leyes que regulan la práctica. Sin embargo, las estadísticas relacionadas con accidentes en la construcción han permanecido casi constantes desde 1985 al presente. Por esta razón, se hace necesario seguir estudiando el problema desde otros abordajes que puedan influir en su mitigación. En el presente trabajo, a través de la argumentación crítica se encuentra una relación estrecha entre dos variables no consideradas hasta el momento de manera conjunta (seguridad vs toma de decisiones sin tener como prioridad consideraciones éticas) y se plantea a la educación en ética como herramienta para crear conciencia y establecer las acciones que hasta ahora han sido evadidas por los agentes de un proyecto de construcción.