Civil & Environmental Engineering
Permanent URI for this communityhttp://hdl.handle.net/1903/2221
Browse
31 results
Search Results
Item REAL-TIME DISPATCHING AND REDEPLOYMENT OF HETEROGENEOUS EMERGENCY VEHICLES FLEET WITH A BALANCED WORKLOAD(2023) Fang, Chenyang; Haghani, Ali; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The emergency management service (EMS) system is a complicated system that tries to coordinate each system component to provide a quick response to emergencies. Different types of vehicles cooperate to finish the tasks under unified command. The EMS system tries to respond quickly to emergency calls and communicate with each department to balance the resources and provide maximal coverage for the whole system. This work aims to develop a highly efficient model for the EMS system to assist the coordinator in making the dispatching and relocation decisions simultaneously. Meanwhile, the model will make a route decision to provide the vehicle drivers with route guidance. In the model, heterogenous emergency vehicle fleets consisting of police vehicles, Basic Life Support (BLS) vehicles, Advanced Life Support (ALS) vehicles, Fire Engines, Fire Trucks, and Fire Quants are considered. Moreover, a coverage strategy is proposed, and different coverage types are considered according to the division of vehicle function. The model tries to provide maximal coverage by advanced vehicles under the premise of ensuring full coverage by basic vehicles. The workload balance of the vehicle crews is considered in the model to ensure fairness. A mathematical model is proposed, then a numerical study is conducted to test the model's performance. The results show that the proposed model can perform well in large-scale problems with significant demands. A comprehensive analysis is conducted on the real-case historical medical data. Then a discrete event simulation system is built. The framework of a discrete event simulation model can mimic the evolution of the entire operation of an emergency response system over time. Finally, the proposed model and discrete event simulation system are applied to the real-case historical medical data. Three different categories of performance measurements are collected, analyzed, and compared with the real-case data. A comprehensive sensitivity analysis is conducted to test the ability of the model to handle different situations. The final results illustrate that the proposed model can improve overall performance in various evaluation metrics.Item A NOVEL APPLICATION OF SELECT AGILE CONCEPTS AND STOCHASTIC ANALYSIS FOR THE OPTIMIZATION OF TRAINING PROGRAMS WITHIN HIGH RELIABILITY ORGANIZATIONS IN HIGH TURN-OVER ENVIRONMENTS AT EDUCATIONAL INSTITUTES AND IN INDUSTRY(2023) Blanton, Richard L; Cui, Qingbin; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)High-turnover environments have been extensively studied with the bulk of the literature focusing on the negative effects on business operations.[1] They present challenges to the resilience of the organization while also limiting the potential profitability from consistently having to spend time training new staff. Furthermore, in manufacturing environments inexperienced staff are prone to mistakes and uncertainty, which can lead to increases in scrap materials and lower production rates due to a lack of mastery of the process. From an organizational standpoint a high-turnover environment presents an unmitigated risk to the organization from the continuous loss of institutional knowledge. This loss can present challenges to the organization in numerous ways, such as capital equipment that no longer has staff qualified or experienced enough to use it leading to costly retraining by the manufacturer or increased risk of a catastrophic failure resulting in damage to the equipment and or injury to the staff. Furthermore, the loss of institutional history leads to the loss of why operations are performed a certain way. As the common saying goes, “those who forget history are bound to repeat it.” which can lead to substantial costs for the organization while old solutions that were previously rejected due to lack of merit are constantly rehashed due to a lack of understanding of how the organization arrived at its current policies. This thesis presents a novel framework to mitigate the potential loss of institutional knowledge via a multifaceted approach. To achieve this a specific topic was identified and used to frame questions that guided the research. Mitigation of the negative impacts of high-turnover in manufacturing environments with a specific focus on the optimization of training programs. This topic led to the formulation of the following research questions. What steps can be taken to reduce the chance of lost institutional knowledge in a high-turnover environment? What steps can be taken to reduce the time needed to train a high performing replacement employee, while maintaining strict performance and safety standards? What steps should be taken to improve the accuracy of budgetary projections? What steps need to be taken to enable accurate analysis of potential future investment opportunities in a training program. The answers to the above research questions are compiled and presented with the aim to provide professionals, who are responsible for training programs in high-turnover environments that require a high organizational reliability, with a framework and analysis toolset that will enable data-driven decision making regarding the program. Additionally this thesis provides a framework for addressing the continuous risk of loss of institutional knowledge. When contrasted with a standard training model, where a trainee is presented with new material and then tested for retention before moving to the next topic, the proposed model implements a schema that can be rapidly iterated upon and improved until the desired performance outcome is achieved, while increasing the potential accuracy of budgetary estimation by as much as 57%. Throughout the process, decision makers will have insight into the long term effects of their potential actions by way of running simulations that give insight into not only the expected steady-state cost of a program but also the rough volume of trainees required to achieve that steady-state.Item EVENT-DRIVEN OPERATION OF DISTRIBUTED SYSTEMS WITH ARTIFICIAL INTELLIGENCE TECHNOLOGIES AND BEHAVIOR MODELING(2022) Montezzo Coelho, Maria Eduarda; Austin, Mark A; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This dissertation aims to enhance decision making in urban settings by integrating artificial intelligence technologies with distributed behavior modeling. Today’s civil engineering systems are far more heterogeneous than their predecessors and may be connected to other types of systems in completely new ways, making the task of system design, analysis and integration of multi-disciplinary concerns much more difficult than in the past. These challenges can be addressed by combining machine learning formalisms and semantic model representations of urban systems, that work side-by-side in collecting data, identifying events, and managing city operations in real-time. We exercise the proposed approach on a problem involving anomaly detection in an urbanwater distribution system and a metrorail system.Item INTRODUCING MEETING LOCATIONS IN TAXI-SHARING SYSTEMS(2021) Aliari, Sanaz; Haghani, Ali A.H.; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Taxi-sharing is admittedly a promising solution to reduce travel costs and mitigate congestion. However, sharing a ride with other passengers may impose long detours. Taxi-sharing can be more beneficial if the vehicle miles traveled for the detours can be reduced when serving additional passengers. In this study, we propose incorporating alternative meeting points in taxi-sharing routing to further boost efficiency by eliminating unnecessary detours and improving the chances of passengers being matched. Unlike traditional taxi-sharing systems, passengers are not necessarily picked up at their original location in the proposed system. Instead, they are serviced within a walkable distance of their origin (meeting points). This can be especially beneficial in heavily congested road networks in dense urban areas.There are several challenges in the real-world implementation of a ride-sharing system with alternative pick-up points in a dense high-demand network. The first challenge is to find appropriate passenger matches that may share a ride. The second challenge is to effectively find a set of specific locations as the potential alternative pick-up points that are likely to reduce total travel times. Once possible matches and the corresponding set of candidate pickup points are selected, the last challenge is to obtain optimal/near-optimal routes to satisfy the passengers’ demand with a reasonable computational time. In this study, we formulate a mixed integer programming model for the ride-sharing problem with alternative pick-up points and introduce strategies to cope with these challenges in a real-world setting. We use the New York City taxi demand data that may sometimes have hundreds of demands per minute in a relatively small geographical area. We first introduce a decomposition procedure to break down such a large-scale problem into smaller subproblems by pre-matching groups of passengers that are more likely to share a ride. We then create an initial set of candidate locations for all pick-up points in each group. Then, different strategies are proposed to reduce the set of candidate locations into smaller subsets, including alternative locations with higher potentials of constructing better routes. The experimental results show that incorporating alternative pick-up points results in significant savings in total travel times, total wait times, and the number of vehicles used compared to a baseline ride-sharing system that picks up each passenger exactly from their requested location. Finally, given the high computational cost of the optimization problem, we propose a genetic algorithm to find near-optimal solutions for the formulated problem, and we show that the proposed algorithm achieves solutions that are very close to the optimal solutions, with significantly lower computational times. We design specific operations to implement basic components of the genetic algorithm in the context of our algorithm. This includes strategies to diversify and create random sets of feasible solutions (individuals), select the fittest individuals in each generation, and create new generations through mutations and ordered cross-over such that the new individuals can inherit from their parents while still representing a feasible solution.Item FLIGHT ARRIVAL SCHEDULING MODELS FOR INCORPORATING COLLABORATIVE DECISION-MAKING CONCEPTS INTO TIME-BASED FLOW MANAGEMENT(2021) HAO, YEMING; Lovell, David J.; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Time-based air traffic flow management balances demand versus capacity at facilities by assigning Controlled Times of Arrival (CTAs) to incoming flights. However, existing systems do not differentiate delay costs among different aircraft and do not take user preferences into consideration. From a business perspective, it is essential to understand user preferences and to allow users to engage in decision-making. This dissertation presents results of simulations of strategies to incorporate business-driven airline preferences into these air traffic flow management systems following a Collaborative Decision-Making paradigm. We evaluate optimization models and heuristics to assign CTAs based on user-provided information and priority preferences in a way that minimizes the total CTA delay cost. Potential savings were quantified by comparing the results with the default first-come-first-served (FCFS) scheme. Monte Carlo simulations are conducted using historical flights data under a variety of realistic scenarios. Results show that our proposed heuristic, 2OptSwap, could reduce CTA delay cost between 20% and 30% relative to the FCFS baseline scheme, maintaining an average of 5.9% gap compared to the optimal solution. It is also shown that the heuristic could potentially realize the same level of cost savings regardless of whether the incoming flights are behind their schedules or not. Additional findings include that by starting the flow management further away, we could achieve more delay cost savings. The environment in the air space (such as the wind and the facility capacity) is constantly changing. Therefore, a rolling horizon approach was integrated into this air traffic flow management system. This approach allows the system to incorporate the most recent information at each epoch and solve the problem in a dynamic fashion. Real-time fairness metrics and adjustment methods are defined such that performance measurements in the previous epochs can be used for adjustments in future decision-making. Simulation results show that these fairness adjustments can help achieve a fairer benefits distribution among carriers and achieve a win-win solution.Item Dynamic Repositioning For Bikesharing Systems(2020) Roshan Zamir, Kiana; Haghani, Ali; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Bikesharing systems’ popularity has continuously been rising during the past years due to technological advancements. Managing and maintaining these emerging systems are indispensable parts of these systems and are necessary for their sustainable growth and successful implementation. One of the challenges that operators of these systems are facing is the uneven distribution of bikes due to users’ activities. These imbalances in the system can result in a lack of bikes or docks and consequently cause user dissatisfaction. A dynamic repositioning model that integrates prediction and routing is proposed to address this challenge. This operational model includes prediction, optimization, and simulation modules and can assist the operators of these systems in maintaining an effective system during peak periods with less number of unmet demands. It also can provide insights for planners by preparing development plans with the ultimate goal of more efficient systems. Developing a reliable prediction module that has the ability to predict future station-level demands can help system operators cope with the rebalancing needs more effectively. In this research, we utilize the expressive power of neural networks for predicting station-level demands (number of pick-ups and drop-offs) of bikeshare systems over multiple future time intervals. We examine the possibility of improving predictions by taking into account new sources of information about these systems, namely membership type and status of stations. A mathematical formulation is then developed for repositioning the bikes in the system with the goal of minimizing the number of unmet demands. The proposed module is a dynamic multi-period model with a rolling horizon which accounts for demands in the future time intervals. The performance of the optimization module and its assumptions are evaluated using discrete event simulation. Also, a three-step heuristic method is developed for solving large-size problems in a reasonable time. Finally, the integrated model is tested on several case studies from Capital Bikeshare, the District of Columbia’s bikeshare program.Item DEVELOPMENT OF AN INTEGRATED RIDE-SHARED MOBILITY-ON-DEMAND (MOD) AND PUBLIC TRANSIT SYSTEM(2019) XU, LIU; Haghani, Ali; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The Mobility-on-Demand (MOD) services, like the ones offered by Uber and Lyft, are transforming urban transportation by providing more sustainable and convenient service that allows people to access anytime and anywhere. In most U.S. cities with sprawling suburban areas, the utilization of public transit for commuting is often low due to lack of accessibility. Thereby the MOD system can function as a first-and-last-mile solution to attract more riders to use public transit. Seamless integration of ride-shared MOD service with public transit presents enormous potential in reducing pollution, saving energy, and alleviating congestion. This research proposes a general mathematical framework for solving a multi-modal large-scale ride-sharing problem under real-time context. The framework consists of three core modules. The first module partitions the entire map into a set of more scalable zones to enhance computational efficiency. The second module encompasses a mixed-integer-programming model to concurrently find the optimal vehicle-to-request and request-to-request matches in a hybrid network. The third module forecasts the demand for each station in the near future and then generates an optimized vehicle allocation plan to best serve the incoming rider requests. To ensure its applicability, the proposed model accounts for transit frequency, MOD vehicle capacity, available fleet size, customer walk-away condition and travel time uncertainty. Extensive experimental results prove that the proposed system can bring significant vehicular emission reduction and deliver timely ride-sharing service for a large number of riders. The main contributions of this study are as follows: • Design of a general framework for planning a multi-modal ride-sharing system in cities with under-utilized public transit system; • Development of an efficient real-time algorithm that can produce solutions of desired quality and scalability and redistribute the available fleet corresponding to the future demand evolution; • Validation of the potential applicability of the proposed system and quantitatively reveal the trade-off between service quality and system efficiency.Item Solving the integrated school bell time, and bus routing and scheduling optimization problem under the deterministic and stochastic conditions(2019) Wang, Zhongxiang; Haghani, Ali; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The school bus planning problem (SBPP) has drawn significant attention in research and practice because of its importance in pupil transportation. The major task of the SBPP is to simultaneously optimize the school bell times, the routing plan (a set of trips) and the scheduling plan (the assignment of buses to serve these trips) while maintaining the minimum level-of-service requirements with the objective that the total number of buses and the total vehicle time are both minimized. Many subproblems of the SBPP have been well studied, but the integrated problem lacks much research due to its complexity. A Mixed Integer Programming (MIP) model is proposed for the integrated SBPP. A novel decomposition method is developed to solve the model. It distinguishes itself from the literature with the consideration of trip compatibility in the routing stage, which is a piece of essential information in the following scheduling stage. This ‘look ahead’ strategy finds a new balance between the model integration and decomposition, which solves the problem efficiently as a decomposed problem but with the high solution quality as the integrated model. Three heuristic algorithms are proposed to solve the deterministic SBPP with the trip compatibility. Then, two mathematical programming models and a Column Generation-based algorithm are proposed for the SBPP under traffic congestion and stochastic travel time in a real uncertain world. These innovative algorithms incorporate the merits of the Simulated Annealing, Tabu Search, Insertion Algorithm, and Greedy Randomized Adaptive Search Procedure and gain the computational power that the existing methods do not have. The experiments are conducted on randomly generated datasets, benchmark problems, and real-world cases. The results show that the proposed models and algorithms outperform the state-of-the-art method in all test problems by up to 25%. In a real-world case study, after the bell time adjustment, up to 41% of current buses can be saved with even better service with respect to the higher punctuality and shorter student ride time.Item Optimal Reassignment of Flights to Gates Focusing on Transfer Passengers(2019) Pternea, Moschoula; Haghani, Ali; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This dissertation focuses on the optimal flight-to-gate assignment in cases of schedule disruptions with a focus on transfer passengers. Disruptions result from increased passenger demand, combined with tight scheduling and limited infrastructure capacity. The critical role of gate assignment, combined with the scarcity of models and algorithms to handle passenger connections, is the main motivation for this study. Our first task is to develop a generalizable multidimensional assignment model that considers the location of gates and the required connection time to assess the success of passenger transfers. The results demonstrate that considering gate location is critical for assessing of the success of a connection, since transfer passengers contribute significantly to total cost. We then explore the mathematical programming formulation of the problem. First, we compare different state-of-art mathematical formulations, and identify their underlying assumptions. Then, we strengthen our time-index formulation by introducing valid inequalities. Afterwards, we express the cost of passenger connections using an aggregating formulation, which outperforms the quadratic formulation and is consistently more efficient than network flow formulations when the cost of successful connections is considered. In the last part of the dissertation, we embed the formulation in an MIP-based metaheuristic framework using Variable Neighborhood Search with Local Branching (VNS-LB). We explore the key notion of a solution neighborhood in the context of gate assignment, given that transfer passengers are our main consideration. Our implementation produces near-optimal results in a low amount of time and responds reasonably to sensitivity analysis in operating parameters and external conditions. Furthermore, VNS-LB is shown to outperform the Local Branching heuristic in terms of solution quality. Finally, we propose a set of extensions to the algorithm which are shown to improve the quality of the final solution, as well as the progress of the optimization procedure as a whole. This dissertation aspires to develop a versatile tool that can be adapted to the objectives and priorities of practitioners, and to provide researchers with an insight of how the features of a solution are reflected in the mathematical formulation. Every idea relying on these principles should be a promising path for future research.Item Mechanisms for Trajectory Options Allocation in Collaborative Air Traffic Flow Management(2018) Mohanavelu Umamagesh, Prithiv Raj; Lovell, David J; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Flight delays are primarily due to traffic imbalances caused by the demand for airspace resource exceeding its capacity. The capacity restriction might be due to inclement weather, an overloaded air traffic sector, or an airspace restriction. The Federal Aviation Administration (FAA), the organization responsible for air traffic control and management in the USA, has developed several tools known as Traffic Management Initiatives (TMI) to bring the demand into compliance with the capacity constraints. Collaborative Trajectory Option Program (CTOP) is one such tool that has been developed by the FAA to mitigate the delay experienced by flights. Operating under a Collaborative Decision Making (CDM) environment, CTOP is considered as the next step into the future of air traffic management by the FAA. The advantages of CTOP over the traditional the TMIs are unequivocal. The concerns about the allocation scheme used in the CTOP and treatment of flights from the flight operators/airlines have limited its usage. This research was motivated by the high ground delays that were experienced by flights and how the rerouting decisions were made in the current allocation method used in a CTOP. We have proposed four alternative approaches in this thesis, which incorporated priority of flights by the respective flight operator, aimed at not merely reducing an individual flight operator’s delay but also the total delay incurred to the system. We developed a test case scenario to compare the performances of the four proposed allocation methods against one another and with the present allocation mechanism of CTOP.