ABSTRACT Due to the increasing population and limited funding for maintenance and construction, t... more ABSTRACT Due to the increasing population and limited funding for maintenance and construction, the efficiency of the transportation network system in the U.S. is being challenged by a potential crisis that endangers the economic growth of the nation and the quality of life of the people. Although funding allocation can be approached using different methods, it is difficult to reach consensus on the criteria used to distribute limited funds among competing needs. Conventional funding allocation methods use weighted formulas based on population and other criteria. However, formula-based funding allocation methods may lead to the public's disagreement if funding allocation final decisions are not perceived as fair or equitable by all agencies. A fair division transportation allocation model (FDTAM) is proposed as an alternative to fairly distribute limited funds among the agencies competing for funding. The FDTAM aims to maximize the participants' project desirability and to minimize total envy due to budget constraints. Desirability represents how desirable a certain project is to the participant based on his/her own criteria. As a result of a case study, the application of the FDTAM to transportation allocation problems is considered a feasible alternative when compared with conventional funding allocation methods. (C) 2014 American Society of Civil Engineers.
International Journal of Information Technology Project Management, 2010
Abstract Berth scheduling can be described as the resource allocation problem of berth space to v... more Abstract Berth scheduling can be described as the resource allocation problem of berth space to vessels in a container terminal. When defining the allocation of berths to vessels container terminal operators set several objectives which ideally need to be optimized simultaneously. These multiple objectives are often non-commensurable and gaining an improvement on one objective often causes degrading performance on the other objectives. In this paper, the authors present the application of a multi-objective decision and analysis ...
Multiple objective system reliability optimization problems have been become more popular in engi... more Multiple objective system reliability optimization problems have been become more popular in engineering and have been deeply covered in literature. There are different approaches to address multiple objective optimization problems depending of the area and special characteristics of the problem. Most of the methods developed to solve these types of problems consist in a Pareto set of optimal solutions. At this point the decision-maker has to select one solution among the Pareto set. This task is not trivial due to the large size of the set to choose for. This work presents a new method based on self-organizing trees for reducing the size of Pareto-optimal sets. The method is tested in several Pareto sets from different multiple objective system reliability optimization problems including a network reliability design problem.
Proceedings of The Institution of Mechanical Engineers Part O-journal of Risk and Reliability, 2008
... pj º lim t!1 PG ºgjfi ¬ √ , gj º jth performance level 1fi A º X gj >D pj 2fi In electric ... more ... pj º lim t!1 PG ºgjfi ¬ √ , gj º jth performance level 1fi A º X gj >D pj 2fi In electric power systems, the operation periods T are divided into S intervals with durations T1, T2, ..., TS, and each interval has a required demand level (D1, D2, ..., DS ) respectively. ...
This paper proposes a practical methodology for the solution of multi-objective system reliabilit... more This paper proposes a practical methodology for the solution of multi-objective system reliability optimization problems. The new method is based on the sequential combination of multi-objective evolutionary algorithms and data clustering on the prospective solutions to yield a smaller, more manageable sets of prospective solutions. Existing methods for multiple objective problems involve either the consolidation of all objectives into a
A newly developed multiple objective evolutionary algorithm is presented. MoPriGA, a multi-object... more A newly developed multiple objective evolutionary algorithm is presented. MoPriGA, a multi-objective prioritized genetic algorithm, incorporates the knowledge of the decision-maker objective function preferences directly within the evolutionary algorithm. The idea behind this algorithm is to more intensely focus on the region of the Pareto set of interest to the decision-maker.
ABSTRACT There are often multiple competing objectives for industrial scheduling and production p... more ABSTRACT There are often multiple competing objectives for industrial scheduling and production planning problems. Two practical methods are presented to efficiently identify promising solutions from among a Pareto-optimal set for multiple objective scheduling problems. Generally, multi-objective optimization problems can be solved either by combining the objectives into a single objective function using equivalent cost conversions, utility theory, etc., or by determination of a Pareto-optimal set. Pareto-optimal sets or representative sub-sets can be found by using a multi-objective genetic algorithm or by other means. Then, in practice, the decision-maker ultimately has to select one solution from this set for system implementation. However, the Pareto-optimal set is often large and cumbersome, making the post-Pareto analysis phase potentially difficult, especially as the number of objectives increase. Our research involves the post-Pareto analysis phase, and two methods are presented to filter the Pareto-optimal set to determine a subset of promising or desirable solutions. The first method is pruning using non- numerical objective function ranking preferences. The second approach involves pruning by using data clustering. The k-means algorithm is used to find clusters of similar solutions in the Pareto-optimal set. The clustered data allows the decision maker to have just k general solutions to choose from. These methods are general, and they are demonstrated using a multi-objective problem involving the scheduling of the bottleneck operation of a printed wiring board manufacturing line.
A new multiple objective evolutionary algorithm is proposed for solving system design allocation ... more A new multiple objective evolutionary algorithm is proposed for solving system design allocation problems. The developed algorithm mainly differs from other MOEAs in the crossover operation performed and in the fitness assignment. In the crossover step, several offspring are created through multi-parent recombination. Thus, the mating pool contains a great amount of diversity of solutions. This disruptive nature of our proposed type of crossover, called subsystem rotation crossover (SURC) encourages the exploration of the search space. The algorithm was thoroughly tested and a performance comparison of the proposed algorithm against one of the most successful MOEAs that currently exists shows that our algorithm is more powerful to solve multi-objective redundant design allocation problems. 1. Introduction This paper describes the use of a multiple objective evolutionary algorithm to solve engineering design allocation problems. The problem addressed in the paper arises in many real engineering optimization problems, where managers and/or decision-makers have to efficiently allocate components from among of a set of predefined component choices to determine the optimal configuration to be implemented. There are numerous application areas of the redundancy allocation problem, such as in the case of electrical power systems, transportation systems, and telecommunications among others (Levitin & Lisnianski, 2001; Lyu et al ., 2002, etc.). This paper addresses the problem of designing a hardware system structure. In the problem formulation presented, there is a specified number of subsystems and, for each subsystem, there are multiple component choices which can be selected and used in parallel. This formulation pertains to the well-known redundancy allocation problem (RAP). In this paper, the RAP is modeled as a multi-objective problem with the system reliability to be maximized, cost and weight of the system to be minimized, and no constraints limiting the possible values of reliability, making this problem a multiple objective combinatorial optimization (MOCO) problem.
ABSTRACT Two methods are presented as practical approaches to reduce the size of the Pareto optim... more ABSTRACT Two methods are presented as practical approaches to reduce the size of the Pareto optimal set of multiple-objective system reliability design problems. The first method is a pseudo-ranking scheme that helps the decision-maker select solutions that reflect his/her preferences. In the second approach we used clustering techniques used in data mining, to group the data by using the k-means algorithm to find clusters of similar solutions, which allows the decision-maker to have just k solutions to choose from without using any objective function preference information. Under this second method, from the clustered Pareto optimal set, we attempted to find solutions which are likely to be more relevant to the decision-maker, which are solutions where a small improvement in one objective would lead to a large deterioration in at least one other objective. To show how these methods work, the well-known Redundancy Allocation Problem was solved as a multiple objective problem by using the NSGA genetic algorithm to initially find the Pareto optimal solutions, and then, the two proposed methods are applied to prune the Pareto set.
ABSTRACT There are often multiple competing objectives for industrial scheduling and production p... more ABSTRACT There are often multiple competing objectives for industrial scheduling and production planning problems. Two practical methods are presented to efficiently identify promising solutions from among a Pareto-optimal set for multiple objective scheduling problems. Generally, multi-objective optimization problems can be solved either by combining the objectives into a single objective function using equivalent cost conversions, utility theory, etc., or by determination of a Pareto-optimal set. Pareto-optimal sets or representative sub-sets can be found by using a multi-objective genetic algorithm or by other means. Then, in practice, the decision-maker ultimately has to select one solution from this set for system implementation. However, the Pareto-optimal set is often large and cumbersome, making the post-Pareto analysis phase potentially difficult, especially as the number of objectives increase. Our research involves the post-Pareto analysis phase, and two methods are presented to filter the Pareto-optimal set to determine a subset of promising or desirable solutions. The first method is pruning using non- numerical objective function ranking preferences. The second approach involves pruning by using data clustering. The k-means algorithm is used to find clusters of similar solutions in the Pareto-optimal set. The clustered data allows the decision maker to have just k general solutions to choose from. These methods are general, and they are demonstrated using a multi-objective problem involving the scheduling of the bottleneck operation of a printed wiring board manufacturing line.
... For instance, consider the following example to illustrate a chromosome used in our algorithm... more ... For instance, consider the following example to illustrate a chromosome used in our algorithm to solve the multiobjective-RAP problem. ... infeasible, regions in an efficient way. ... difficult to implement, because the exact location of the boundary between the feasible and ...
ABSTRACT Due to the increasing population and limited funding for maintenance and construction, t... more ABSTRACT Due to the increasing population and limited funding for maintenance and construction, the efficiency of the transportation network system in the U.S. is being challenged by a potential crisis that endangers the economic growth of the nation and the quality of life of the people. Although funding allocation can be approached using different methods, it is difficult to reach consensus on the criteria used to distribute limited funds among competing needs. Conventional funding allocation methods use weighted formulas based on population and other criteria. However, formula-based funding allocation methods may lead to the public's disagreement if funding allocation final decisions are not perceived as fair or equitable by all agencies. A fair division transportation allocation model (FDTAM) is proposed as an alternative to fairly distribute limited funds among the agencies competing for funding. The FDTAM aims to maximize the participants' project desirability and to minimize total envy due to budget constraints. Desirability represents how desirable a certain project is to the participant based on his/her own criteria. As a result of a case study, the application of the FDTAM to transportation allocation problems is considered a feasible alternative when compared with conventional funding allocation methods. (C) 2014 American Society of Civil Engineers.
International Journal of Information Technology Project Management, 2010
Abstract Berth scheduling can be described as the resource allocation problem of berth space to v... more Abstract Berth scheduling can be described as the resource allocation problem of berth space to vessels in a container terminal. When defining the allocation of berths to vessels container terminal operators set several objectives which ideally need to be optimized simultaneously. These multiple objectives are often non-commensurable and gaining an improvement on one objective often causes degrading performance on the other objectives. In this paper, the authors present the application of a multi-objective decision and analysis ...
Multiple objective system reliability optimization problems have been become more popular in engi... more Multiple objective system reliability optimization problems have been become more popular in engineering and have been deeply covered in literature. There are different approaches to address multiple objective optimization problems depending of the area and special characteristics of the problem. Most of the methods developed to solve these types of problems consist in a Pareto set of optimal solutions. At this point the decision-maker has to select one solution among the Pareto set. This task is not trivial due to the large size of the set to choose for. This work presents a new method based on self-organizing trees for reducing the size of Pareto-optimal sets. The method is tested in several Pareto sets from different multiple objective system reliability optimization problems including a network reliability design problem.
Proceedings of The Institution of Mechanical Engineers Part O-journal of Risk and Reliability, 2008
... pj º lim t!1 PG ºgjfi ¬ √ , gj º jth performance level 1fi A º X gj >D pj 2fi In electric ... more ... pj º lim t!1 PG ºgjfi ¬ √ , gj º jth performance level 1fi A º X gj >D pj 2fi In electric power systems, the operation periods T are divided into S intervals with durations T1, T2, ..., TS, and each interval has a required demand level (D1, D2, ..., DS ) respectively. ...
This paper proposes a practical methodology for the solution of multi-objective system reliabilit... more This paper proposes a practical methodology for the solution of multi-objective system reliability optimization problems. The new method is based on the sequential combination of multi-objective evolutionary algorithms and data clustering on the prospective solutions to yield a smaller, more manageable sets of prospective solutions. Existing methods for multiple objective problems involve either the consolidation of all objectives into a
A newly developed multiple objective evolutionary algorithm is presented. MoPriGA, a multi-object... more A newly developed multiple objective evolutionary algorithm is presented. MoPriGA, a multi-objective prioritized genetic algorithm, incorporates the knowledge of the decision-maker objective function preferences directly within the evolutionary algorithm. The idea behind this algorithm is to more intensely focus on the region of the Pareto set of interest to the decision-maker.
ABSTRACT There are often multiple competing objectives for industrial scheduling and production p... more ABSTRACT There are often multiple competing objectives for industrial scheduling and production planning problems. Two practical methods are presented to efficiently identify promising solutions from among a Pareto-optimal set for multiple objective scheduling problems. Generally, multi-objective optimization problems can be solved either by combining the objectives into a single objective function using equivalent cost conversions, utility theory, etc., or by determination of a Pareto-optimal set. Pareto-optimal sets or representative sub-sets can be found by using a multi-objective genetic algorithm or by other means. Then, in practice, the decision-maker ultimately has to select one solution from this set for system implementation. However, the Pareto-optimal set is often large and cumbersome, making the post-Pareto analysis phase potentially difficult, especially as the number of objectives increase. Our research involves the post-Pareto analysis phase, and two methods are presented to filter the Pareto-optimal set to determine a subset of promising or desirable solutions. The first method is pruning using non- numerical objective function ranking preferences. The second approach involves pruning by using data clustering. The k-means algorithm is used to find clusters of similar solutions in the Pareto-optimal set. The clustered data allows the decision maker to have just k general solutions to choose from. These methods are general, and they are demonstrated using a multi-objective problem involving the scheduling of the bottleneck operation of a printed wiring board manufacturing line.
A new multiple objective evolutionary algorithm is proposed for solving system design allocation ... more A new multiple objective evolutionary algorithm is proposed for solving system design allocation problems. The developed algorithm mainly differs from other MOEAs in the crossover operation performed and in the fitness assignment. In the crossover step, several offspring are created through multi-parent recombination. Thus, the mating pool contains a great amount of diversity of solutions. This disruptive nature of our proposed type of crossover, called subsystem rotation crossover (SURC) encourages the exploration of the search space. The algorithm was thoroughly tested and a performance comparison of the proposed algorithm against one of the most successful MOEAs that currently exists shows that our algorithm is more powerful to solve multi-objective redundant design allocation problems. 1. Introduction This paper describes the use of a multiple objective evolutionary algorithm to solve engineering design allocation problems. The problem addressed in the paper arises in many real engineering optimization problems, where managers and/or decision-makers have to efficiently allocate components from among of a set of predefined component choices to determine the optimal configuration to be implemented. There are numerous application areas of the redundancy allocation problem, such as in the case of electrical power systems, transportation systems, and telecommunications among others (Levitin & Lisnianski, 2001; Lyu et al ., 2002, etc.). This paper addresses the problem of designing a hardware system structure. In the problem formulation presented, there is a specified number of subsystems and, for each subsystem, there are multiple component choices which can be selected and used in parallel. This formulation pertains to the well-known redundancy allocation problem (RAP). In this paper, the RAP is modeled as a multi-objective problem with the system reliability to be maximized, cost and weight of the system to be minimized, and no constraints limiting the possible values of reliability, making this problem a multiple objective combinatorial optimization (MOCO) problem.
ABSTRACT Two methods are presented as practical approaches to reduce the size of the Pareto optim... more ABSTRACT Two methods are presented as practical approaches to reduce the size of the Pareto optimal set of multiple-objective system reliability design problems. The first method is a pseudo-ranking scheme that helps the decision-maker select solutions that reflect his/her preferences. In the second approach we used clustering techniques used in data mining, to group the data by using the k-means algorithm to find clusters of similar solutions, which allows the decision-maker to have just k solutions to choose from without using any objective function preference information. Under this second method, from the clustered Pareto optimal set, we attempted to find solutions which are likely to be more relevant to the decision-maker, which are solutions where a small improvement in one objective would lead to a large deterioration in at least one other objective. To show how these methods work, the well-known Redundancy Allocation Problem was solved as a multiple objective problem by using the NSGA genetic algorithm to initially find the Pareto optimal solutions, and then, the two proposed methods are applied to prune the Pareto set.
ABSTRACT There are often multiple competing objectives for industrial scheduling and production p... more ABSTRACT There are often multiple competing objectives for industrial scheduling and production planning problems. Two practical methods are presented to efficiently identify promising solutions from among a Pareto-optimal set for multiple objective scheduling problems. Generally, multi-objective optimization problems can be solved either by combining the objectives into a single objective function using equivalent cost conversions, utility theory, etc., or by determination of a Pareto-optimal set. Pareto-optimal sets or representative sub-sets can be found by using a multi-objective genetic algorithm or by other means. Then, in practice, the decision-maker ultimately has to select one solution from this set for system implementation. However, the Pareto-optimal set is often large and cumbersome, making the post-Pareto analysis phase potentially difficult, especially as the number of objectives increase. Our research involves the post-Pareto analysis phase, and two methods are presented to filter the Pareto-optimal set to determine a subset of promising or desirable solutions. The first method is pruning using non- numerical objective function ranking preferences. The second approach involves pruning by using data clustering. The k-means algorithm is used to find clusters of similar solutions in the Pareto-optimal set. The clustered data allows the decision maker to have just k general solutions to choose from. These methods are general, and they are demonstrated using a multi-objective problem involving the scheduling of the bottleneck operation of a printed wiring board manufacturing line.
... For instance, consider the following example to illustrate a chromosome used in our algorithm... more ... For instance, consider the following example to illustrate a chromosome used in our algorithm to solve the multiobjective-RAP problem. ... infeasible, regions in an efficient way. ... difficult to implement, because the exact location of the boundary between the feasible and ...
Uploads
Papers by Heidi Taboada