A. James Clark School of Engineering
Permanent URI for this communityhttp://hdl.handle.net/1903/1654
The collections in this community comprise faculty research works, as well as graduate theses and dissertations.
Browse
4 results
Search Results
Item INCREASED INDIVIDUAL SIZE AND ITS POTENTIAL EFFECTS ON EMERGENCY EVACUATION SCENARIOS(2018) Ahrens, Katherine; Milke, James; Fire Protection Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The increase in human body size due to obesity and overweight conditions is recognized as becoming more prevalent throughout the world. The effect which increased body size and weight has on movement has been examined from a kinesiological and physiological standpoint. Its effect on egress during emergency evacuation has largely remained unstudied. This study reviews current data on body size using modeling software to examine the potential impact an increase in body size has on evacuation times and whether that impact is significant enough to warrant potential changes to current code and regulatory requirements. The change in body size distribution is analyzed and tests are conducted at increasing body size intervals of 0.025 meters for six different scenarios. Results indicate that an increase of 0.225 meters to a body radius increases evacuation times in simple scenarios between 12% and 72%.Item EVACUATION ROUTE MODELING AND PLANNING WITH GENERAL PURPOSE GPU COMPUTING(2014) Prentiss, David D.; Miller-Hooks, Elise; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This work introduces a bilevel, stochastic optimization problem aimed at robust, regional evacuation network design and shelter location under uncertain hazards. A regional planner, acting as a Stackelberg leader, chooses among evacuation-route contraflow operation and shelter location to minimize the expected risk exposure to evacuees. Evacuees then seek an equilibrium with respect to risk exposure in the lower level. An example network is solved exactly with a strategy that takes advantage of a fast, low-memory, equilibrium algorithm and general purpose computing on graphical processing units.Item OBSERVED BEHAVIOR OF PLATOON DYNAMICS DURING HIGH-RISE STAIRWELL EVACUATIONS(2012) Baker, Matthew Daniel; Milke, James A; Fire Protection Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This research analyzes the phenomena of grouping or platooning during the evacuation of seven stairwells within four different high-rise buildings. The purpose of this research is to investigate the changes occurring to platoons as they descend the stairs in order to incorporate the results into computer egress models. Platoons are found to travel in three distinct patterns: elongation, compression, and equilibrium. Also, platoons are found to remain unchanged, add new occupants, merge with other platoons, or fragment during their descent within a stairwell. The results demonstrate that a trend exists between patterns of platoon elongation leading to fragmentation and platoon compression leading to platoons merging. The majority of the platoons identified are found to consist of one person and remain unchanged as they descend between floors. Finally, a qualitative comparison between the platoons analyzed and the platoons identified in the behavioral computer egress model Pathfinder, is presented.Item On finding paths and flows in multicriteria, stochastic and time-varying networks(2004-11-24) Opasanon, Sathaporn -; Miller-Hooks, Elise; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This dissertation addresses two classes of network flow problems in networks with multiple, stochastic and time-varying attributes. The first problem class is concerned with providing routing instructions with the ability to make updated decisions as information about travel conditions is revealed for individual travelers in a transportation network. Three exact algorithms are presented for identifying all or a subset of the adaptive Pareto-optimal solutions with respect to the expected value of each criterion from each node to a desired destination for each departure time in the period of interest. The second problem class is concerned with problems of determining the optimal set of a priori path flows for evacuation in capacitated networks are addressed, where the time-dependent and stochastic nature of arc attributes and capacities inherent in these problems is explicitly considered. The concept of Safest Escape is formulated for developing egress instructions. An exact algorithm is proposed to determine the pattern of flow that maximizes the minimum path probability of successful arrival of supply at the sink. While the Safest Escape problem considers stochastic, time-varying capacities, arc travel times, while time-varying, are deterministic quantities. Explicit consideration of stochastic and time-varying travel times makes the SEscape problem and other related problems significantly more difficult. A meta-heuristic based on the principles of genetic algorithms is developed for determining optimal path flows with respect to several problems in dynamic networks, where arc traversal times and capacities are random variables with probability mass functions that vary with time. The proposed genetic algorithm is extended for use in more difficult, stochastic, time-varying and multicriteria, capacitated networks, for which no exact, efficient algorithms exist. Several objectives may be simultaneously considered in determining the optimal flow pattern: minimize total time, maximize expected flow and maximize the minimum path probability of successful arrival at the sink (the objective of the SEscape problem). Numerical experiments are conducted to assess the performance of all proposed approaches.