In-Jae Jeong

Bi-objective scheduling for reentrant hybrid flow shop using Pareto genetic algorithm

This paper deals with a scheduling problem for reentrant hybrid flowshop with serial stages where each stage consists of identical parallel machines. In a reentrant flowshop, a job may revisit any stage several times. Local-search based Pareto genetic algorithms with Minkowski distance-based crossover operator is proposed to approximate the Pareto optimal solutions for the minimization of makespan and total tardiness in a reentrant hybrid flowshop. The Pareto genetic algorithms are compared with existing multi-objective genetic algorithm, NSGA-II in terms of the convergence to optimal solution, the diversity of solution and the dominance of solution. Experimental results show that the proposed crossover operator and local search are effective and the proposed algorithm outperforms NSGA-II by statistical analysis.

Integrated decision-support system for diagnosis, maintenance planning, and scheduling of manufacturing systems

This paper describes an integrated decision support system to diagnose faults and generate efficient maintenance and production schedules. The proposed integrated system is composed of three modules, namely, the Diagnosis Module, the Maintenance Planning Module, and the Scheduling Module. In the Diagnosis Module, a vector of symptoms is fed into an influence diagram representing the causal relationships of the system. Given an instantiation of the symptom vector, a stochastic sampling algorithm is applied to update the joint-probability distribution associated with the influence diagram and rank the possible causes of the symptoms. The Maintenance Planning Module searches a Look-Up-Table to find maintenance activities to correct the fault. The Scheduling Module determines the job sequences, including a maintenance activity that can minimize makespan, or total completion time. Also, we tested the diagnostic accuracy of the influence diagram and developed a prototype of the integrated system for a scenario of a single pick-and-place machine in the context of electronics assembly systems.

Decision-making and cooperative interaction via coupling agents in organizationally distributed systems

This paper develops a methodology for decision-making in organizationally distributed systems where decision authorities and information are dispersed in multiple organizations. Global performance is achieved through cooperative interaction and partial information sharing among organizations. The information shared among organizations is contrived using modified Lagrangian relaxation techniques. Novel to the methodology is that no single master problem with a global view of the system is required to guide the decision process. Rather, multiple artificial decision entities, termed Coupling Agents, are associated with subsets of coupling constraints. The proposed generic model can be applied to decision-making problems with a variety of mathematical structures. In this paper the methodology is applied to parameter design problems to illustrate the behavior of the proposed methodology in the realm of non-linear optimization.

A distributed scheduling methodology for a two-machine flowshop using cooperative interaction via multiple coupling agents

This paper presents a distributed scheduling methodology for a two-machine flowshop problem. It is assumed that the decision authorities and information are distributed in multiple subproduction systems that must share two machines to satisfy their demands. The associated scheduling problems are modeled using 0/1 integer formulations, and the problem is solved using Lagrangian relaxation techniques modified to work in an environment where very limited information sharing is allowed. Specifically, no global upper-bound is known, no single decision entity has complete view of all the constraints that couple the participating subproduction systems, and there is no disclosure of local objectives and constraints. The main objective of the proposed algorithm is to find a compromise state where all coupling constraints and local constraints are satisfied, and the total sum of weighted completion time of jobs is minimized. The proposed methodology showed promising experimental results when compared to the traditional Lagrangian relaxation with subgradient method.

Regression approximation for a partially centralized inventory system considering transportation costs

Inventory centralization for multiple stores with stochastic demands reduces costs by establishing and maintaining a central ordering/distribution point. However the inventory centralization may increase the transportation costs since either the customer must travel more to reach the product, or the central warehouse must ship the product over longer distance to reach the customer. In this paper, we study a partially centralized inventory system where multiple central warehouses exist and a central warehouse fulfills the aggregated demand of stores. We want to determine the number, the location of central warehouses and an assignment of central warehouses and a set of stores. The objective is the minimization of the sum of warehouse costs and transportation cost. With the help of the regression approximation of cost function, we transform the original problem to more manageable facility location problems. Regression analysis shows that the approximated cost function is close to the original one for normally distributed demands.

A single-machine distributed scheduling methodology using cooperative interaction via coupling agents

This paper considers a single-machine scheduling problem where the decision authorities and information are distributed in multiple subproduction systems. Subproduction systems share the single machine and must cooperate with one another to achieve a global goal of minimizing a linear function of the completion times of the jobs; e.g., total weighted completion times. It is assumed that neither the subproduction systems nor the shared machine have complete information about the entire system. The associated scheduling problems are formulated as zero-one integer programs. The solution approach is based on Lagrangian relaxation techniques modified to require less global information. Specifically, there is no need for a global upper bound, or a single master problem that has a complete view of all the coupling constraints. The proposed methodology exhibits a promising performance when experimentally compared to the Lagrangian relaxation with a subgradient method with the added benefit that can be applied to situations with more restrictive information sharing.

A job shop distributed scheduling based on Lagrangian relaxation to minimise total completion time

This paper considers a distributed job shop scheduling problem where autonomous sub-production systems share common machines with each other. Each sub-production system is responsible for the scheduling of a set of jobs to minimise the total completion time on shared machines. A sub-production system has ultimate responsibility on maintaining private information such as objective function, processing time and routings on shared machines. Also sub-production systems must cooperate each other in order to achieve a global goal while sharing minimum of private information. In this research, we propose a distributed cooperation method in which sub-production systems and shared machines interact with one another to find a compromised solution between a locally optimised solution and a system-wide solution. We tested the proposed method for small, medium and large size of job shop scheduling problems and compared to a global optimal solutions. The proposed method shows promising results in terms of solution qualities and computational times.

Distributed allocation of the capacity of a single-facility using cooperative interaction via coupling agents

This paper considers the problem associated with the allocation of the finite capacity of a single facility among different business organizations under partial information sharing. In distributed allocation, the decision authorities and system information are dispersed amid organizations and a facility, i.e. no organization requires explicit access to system-wide information in order to allocate effectively the capacity of the shared facility. The lack of explicit information access is compensated by the careful exchange of information among organizations via the shared facility; i.e. cooperative interaction. The facility resolves conflicting interests among organizations on capacity usage and directs locally optimized solutions to a globally optimized solution. The distributed decision making problems associated with each organization and the facility are formulated as linear programs. The proposed cooperative algorithm is tested under two levels of information sharing: when the capacity information of the facility is unknown to organizations, and when partial capacity information of the facility is known to organizations. Experimental results suggest that even in this restricted information environment, the proposed method yields solutions that are comparable to those obtained with methodologies that require unrestricted access to information.

Application of data mining for improving yield in wafer fabrication system

This paper presents a comprehensive and successful application of data mining methodologies to improve wafer yield in a semiconductor wafer fabrication system. To begin with, this paper applies a clustering method to automatically identify AUF (Area Uniform Failure) phenomenon from data instead of visual inspection that bad chips occurs in a specific area of wafer. Next, sequential pattern analysis and classification methods are applied to find out machines and parameters that are cause of low yield, respectively. Finally, this paper demonstrates an information system, Y2R-PLUS (Yield Rapid Ramp-up, Prediction, analysis & Up Support) that is developed in order to analyze wafer yield in a Korea semiconductor manufacturer.

An optimal approach for a set covering version of the refueling-station location problem and its application to a diffusion model

ABSTRACT This article deals with the refueling-station location problem for alternative fuel vehicles in a traffic network. Alternative fuel vehicles can be characterized by the vehicle range that limits the travelable distance with fuel at full capacity. I propose an efficient formulation of the refueling-station location problem using an optimal property and prove that the problem is NP(Non-deterministic Polynomial)-complete in the strong sense. I consider a special case of the refueling-station location problem in which the construction costs are equal for all nodes. In this case, the problem is to determine refueling station locations to minimize the total number of stations, while making the possible multiple predetermined origin–destination round-trips. I propose an optimal algorithm applicable when no refueling stations currently exist in a traffic network and a dynamic programming based algorithm applicable when a set of refueling stations already exists. I apply the algorithms to a traffic network to study the diffusion of refueling stations and predict the speed and range of station establishment. The computational experiments show that the speed of diffusion depends on the vehicle range and the sequence of the origin–destination demands considered in the diffusion process.