- Genetic Programming Theory and Practice III: 9 Pdf
- Table of contents
- Genetic Programming Theory and Practice X
- About this book
- A Constraint programming-based genetic algorithm for capacity output optimization
- Free ebook pdf and epub download directory.
- Combinatorial optimization: Theory and algorithms
- Genetic programming theory and practice x pdf reader
- Working Tips For A Better Ebook Reading Experience
- Machine Learning Control: Genetic Programming
More from Journal of Industrial Engineering and Management. A deterministic algorithm for generating optimal three-stage layouts of A deterministic algorithm for generating optimal three-stage layouts of homogenous strip pieces.
Genetic Programming Theory and Practice III: 9 Pdf
Identity of organizational conflict framework: Evaluating model factors based Identity of organizational conflict framework: Evaluating model factors based on demographic characteristics in Iran. Journal of Industrial Engineering and Management , Nov A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.
A Constraint programming-based genetic algorithm for capacity output optimization , Journal of Industrial Engineering and Management, , pp.
Table of contents
Toggle navigation. More from Journal of Industrial Engineering and Management A deterministic algorithm for generating optimal three-stage layouts of A deterministic algorithm for generating optimal three-stage layouts of homogenous strip pieces Identity of organizational conflict framework: Evaluating model factors based Purpose: The manuscript presents an investigation into a constraint programming-based genetic algorithm for capacity output optimization in a back-end semiconductor manufacturing company.
A genetic algorithm model was created in the second stage to optimize capacity output. Three demand scenarios were applied to test the robustness of the proposed algorithm. Findings: CPGA improved both the machine utilization and capacity output once the minimum requirements of a demand scenario were fulfilled.
The constraints and constructed scenarios were therefore industry-specific. Practical implications: Capacity planning in a semiconductor manufacturing facility need to consider multiple mutually influenced constraints in resource availability, process flow and product demand.
The findings prove that CPGA is a practical and an efficient alternative to optimize the capacity output and to allow the company to review its capacity with quick feedback. Introduction Capacity planning aims to minimize the discrepancy between organization capacity and the product demands to optimize revenue.
Genetic Programming Theory and Practice X
Capacity planning in semiconductor industry is extremely challenging due to its product mix, limited capacity of resources and uneven demands.
Consequentially, a semiconductor manufacturing company needs to balance heterogeneous set of products with different required time or resources in production Naughton, Resources typically include machines, labor, money, time, and raw materials.
The resource cost is commonly a considerable production cost in semiconductor industry. Of these resources, machines are the most critical due to their expensive costs and long acquisition periods. A way for the capacity of expensive resources to be highly utilized is through machine sharing.
By improving machine utilization, capacity output is maximized, which in return results in revenue gains to the company. One example is semiconductor assembly and testing back-end BE production industry, in which the equipment cost ranges from twenty thousands to almost a million US dollars.
For an assembly and testing facility with 80 sets of tools, the total amount of capital investment requires approximately 16 million US dollars.
Saving on capacity could result in gain of a few hundred thousand per year. Effectively allocating all existing capacities while considering a variety of constraints and conflicts incurred from the resources is not easy. In literatures, most studies on the capacity allocation in semiconductor industry are focused on single operation in the product mixes Motivated by the foregoing factors, this research develops an efficient approach to maximize capacity output and machine utilization using constraint programming-based genetic algorithm CPGA.
The approach provides a near optimal solution in spite of the changes in product mix order. A real case study was taken from BE production. The rest of the paper is prepared as follows. Section 2 reviews the related work.
About this book
Section 3 describes the case study company. Section 4 presents the research methodology. Section 5 describes the problem formulation. Section 6 describes an implementation of the proposed method.
Section 7 discusses the results of the proposed solution. Section 8 concludes the study. Mathematical models can be further differentiated into linear and nonlinear; deterministic and stochastic; static and dynamic; discrete and continuous; and deductive, inductive, or floating Ugwa, Mathematical modeling translates identified issues or problems within a system, and breaks it down into usable and mathematical formulations.
The mathematical theory and analysis provide another way of looking at the system.
A Constraint programming-based genetic algorithm for capacity output optimization
Wang and Wang proposed a mathematical model to support the decisions regarding simultaneous resource investment and task allocation plan. The model helps in deciding which is the most profitable among the pending orders in each time bucket under demand and technology uncertainty.
Geng, Jiang and Chen proposed a scenariobased stochastic programming model to describe the uncertain capacity based on overall equipment efficiency. Chen and Lu discussed how the stochastic mixed-integer programming model could be used to determine the robust capacity allocation and expansion policy. Swaminathan provided heuristics to find efficient tool procurement plans and test their quality using Lagrangian relaxation.
Phruksaphanrat, Ohsato and Yenradee proposed aggregate Production Planning model to deal with fuzzy demand and variable system capacity. The new model achieves higher flexibility in estimation and better production plan. Many studies have been conducted to find the tool procurement plan by optimizing the tool capacity allocation.
The mathematical-based modeling and exact solution methods are accurate. However, they are usually time consuming due to the complexity of the problems Wang et al.
Free ebook pdf and epub download directory.
The optimality of the solution depends on the problem domain, given that the variables to be considered are limited. The models derived from artificial intelligence and taking advantage of iterative metaheuristic application of controlled randomization on cumulative results.
Genetic algorithm GA is the most popular metaheuristic model to resolve resource-planning problems Wang et al. GA is based on the concept that a population of candidate solutions should be created and then subjected to an evolutionary process to generate the offspring candidate according to the selection criteria Schneider, Zhang developed a heuristic algorithm, which involves the combination of the GAs and primal-dual algorithm of nonlinear programming, to solve capacity planning under uncertain demand and production consumption.
Wang et al.
Abazari, Solimanpur and Sattari applied GA to find the effective solution from the formulated formula to solve the machine loading problem in flexible manufacturing systems.
Li, Jiang and He developed a genetic-algorithm based method to solve a complex equipment-workforce-service planning problem. Since the problem is highly complex, a GA is sought to solve the problem efficiently. Other notable computer models include tabu search, simulated annealing, and ant colony optimization.
Other evolutionary methods used the ant colony optimization and particle swarm optimization. Computer models, despite relative simplicity in programming, do not guarantee optimal solution. They are approximation techniques in which solution qualities are influenced by the representation, parameters, and problem domain. The development of objective function needs to encompass all the factors of interest, particularly the real-life problems. Constraint programming CP is often integrated into computer models, such as GA, to solve the foregoing problems.
GA is a search method to identify the near optimal solution for the objective function. The basic idea in CP is that the user states the constraints requirements in the problem area Rossi, Van Beek, Walsh, Constraints map out how variables in the program must relate to each other. Each variable take a value in a given domain.
Combinatorial optimization: Theory and algorithms
The constraint thus restricts the possible values that variables can take. The important feature of constraints is their declarative manner. They define what relationship must hold within the variables without specifying a computational procedure to enforce the relationship. The quality of solution is usually measured by an application-dependent function referred to as objective function. Van Beek and Chen presented evidence that CP approach can work well in planning and has the advantage in terms of time and space efficiency.
The model integrates three branch and bound methods and considers precedence constraints, capacity constraints, release time and due date. Tang, Liu and Sun integrated the linear scheduling and CP to solve schedule control problems faced during railroad construction.
Genetic programming theory and practice x pdf reader
Case study Company A, a multinational BE semiconductor manufacturing company located in Penang, with approximately 4, employees, was used as the case study. Company A supplies customers from approximately countries worldwide. In , the company achieved a turnover of 5 billion Euro. Company A manufactures products, such as light emitting diodes for automotive, consumer, and industrial applications, infrared products, laser diodes, and optical sensors.
Company A uses lead frame in manufacturing chip packages. The lead frame is in panel form, which is arranged in a matrix that is extended in multiple rows and columns. The total number of units produced from one lead frame panel is different from others based on the size of the chip packages. The lead frame moves through the assembly facilities in a lot collection of lead frame panels. A number of processing steps are performed by the single lead frame panel, while other steps are performed on the entire lot, with several lots processed at the same time.
The whole processing steps are depicted in Figure 1. Several product types with similar process steps are grouped into product families.
Working Tips For A Better Ebook Reading Experience
Machine capacities are predominantly shared among the product family to reduce the investment cost. Setup times for the machine conversion, machine capability, and other mutually influenced constraints therefore are inevitable. Machine capability refers to the technical manufacturing ability of the machine to meet certain product specification.
The production of a product generally involves a series of operations on different machines and cycle times.
While generally the bottleneck operation limits the output of the entire products, a change in the demand for a specific product shifts the bottleneck operation, which affects the output of other products. This study investigates an approach that can maximize both the output and utilization of largely shared machines.
Machine Learning Control: Genetic Programming
The specific manufacturing period ranges from one month to six months.