|author||Kim Hyun Jun|
|2nd author||Jinwoo Park
Sang Hyup Lee
Seung Jin Ha
|presenter||Kim Hyun Jun|
|info||This conference paper has been presented by Kim Hyun Jun the Asia-Simulation Conference. The conference has been held in Jeju, Korea, Republic of, 2015/11/04 ~ 2014/11/07 .|
|start / end date||2015-11-04 ~ 2015-11-07|
|city / nation||Jeju / Korea|
|keywords||Lay-out Planning, Priority Policy, Dynamic Programming, Simulation|
|abstract||Though numerous studies have dealt the lay-out planning methods for shipbuilding industry, in practice the lay-out planning is still done by people under the rule of thumb. It seems illogical to treat such key resource without any scientific management, however matters such as reliability, frequent modifications, and lack of data makes it hard to adapt an automated method in such spatial planning.
To investigate on this problem this paper examines a heavy industry company in Korea and executes a simulation study. The targeted company produces numerous types of offshore plants which all follows varying production processes. So it is told that there exists some special difficulties in predicting and managing newly given constructions. Also another characteristic that makes managing such work difficult is that there exists frequent change requests to design process, quantity, and schedule.
Like many other competitive companies, company H still relies solely on human resource to plan the lay-outs without using any automated systems. This could cause problems such as long hours spent in planning/evaluating the lay-out plan and the difficulty in attaining a substitute for such planning jobs. This is because the current lay-out planning process depends heavily on the experiences of the planner. The difficulty of training a worker to become such specialist is one thing, but to find a substitute in the case of emergency is a more serious problem.
To solve this problem this paper suggests a practical algorithm that can deploy the blocks under a short given time. The system would automatically prioritize the blocks that needs to be placed on the yards and find a location that could maximize the yard utilization and minimize the delays on the site. However because of the uncertainties in the field, such as assembly modifications, contract date delays, and variations in the processing time, it is extremely difficult to find an optimal solution in such short given time.
Studies done by Li(2005) have divided the lay-out process into two separate steps, prescheduling and spatial planning. Here the second step utilized the genetic algorithm. Other researches such as Raj(2007) treated this problem as a 3D bin packing problem, while Zheng(2011) approached this matter using greedy algorithm to minimize the makespan. However none of the previous works considered the uncertainties that exits on the spatial planning process.
So to aid the suggested problem this simulation study aims to find a priority rule to deploy the blocks while considering the uncertainties. Measures such as block size, delayed time of the selected block, and so on would be examined through simulation to find the best performing weights for such variables. Blocks will be prioritized under this rule and will be placed on the yards using the suggested algorithm.
A typical process of such simulation would go through following steps. The first step would be deciding the timeline of deployment. After selecting the timeline target project will be selected. The term project used here is a group of modules and blocks that needs to be placed on the yard. The next step would be choosing the weights for the priority rule. Measures such as delayed time, deployment time, and the size of the block would be considered. By simulating the model while varying the weights of such measures, we will be able to suggest the best performing block selecting policy rule for the spatial lay-out planning.
|75||IC||ICRA||An On-Line Production Scheduler using Neural Network and Simulator based on Manufacturing System State||Kitae Kim||2001|
|74||IC||PEDAC||A Shifting Bottleneck Procedure for Job-Shop Scheduling minimizing the Sum of Earliness and Tardiness||Daeyoung Chung||2001|
|73||IC||ICMA||Machine-Understandable e-Business Modeling with Ontology||Tai-Woo Chang||2002|
|72||IC||Asian eBiz||A Systematic Approach for Ontology-based e-Business Modeling||Tai-Woo Chang||2002|
|71||IC||SCSC||Performance Improvement Methodology for a Manufacturing System using Petri Nets and Simulation Analysis||Kichang Lee||2002|
|70||IC||SCSC||Inventory Management in a Supply Chain using Neural Networks||Sungwon Jung||2002|
|69||IC||IBERC||An Ontology-Based Framework for e-Business Integration||Tai-Woo Chang||2003|
|68||IC||ICPR||Combined production and distribution planning using demand based mixed Genetic Algorithm in a supply chain||Sungwon Jung||2003|
|67||IC||CIE||Decision support system for preliminary cost estimation with case based reasoning||Jonghan Kim||2004|
|66||IC||SCSC||Performance Evaluation of Alternative Designs for the Car Recycling System Using Simulation Analysis||Eoksu Sim||2004|
|65||IC||GECCO||A Generic Network Design for a Closed-loop Supply Chain Using Genetic Algorithm||Eoksu Sim||2004|
|64||IC||AsiaSim||Vendor Managed Inventory and its value in the various supply chain||Sungwon Jung||2004|
|63||IC||AsiaSim||Performance Analysis of Alternative Designs for a Vehicle Disassembly System using Simulation Modeling||Eoksu Sim||2004|
|62||IC||AsiaSim||Reorder decision system based on the concept of the order risk using neural networks||Sungwon Jung||2004|
|61||IC||CIE||Information Strategy Planning for the Korean Postal Address Database||Tai-Woo Chang||2004|
|60||IC||e-Biz||Design of Information System for e-Logistics||Kitae Shin||2004|
|59||IC||ICCSA||Efficient Mapping Rule of IDEF for UMM Application||Kitae Shin||2005|
|58||IC||KGW||A study on due-date assignment method in wafer fabrication facility producing multiple product types||Jaewon Park||2005|
|57||IC||Asian eBiz||Applying Negotiation Patterns in Supply Chain Planning based on Self-integrating Environment||Yuncheol Kang||2005|
|56||IC||GECCO||Solving Multi-level Lot Sizing Problem with Memetic Algorithm based on Refinement Procedure||Haejoong Kim||2006|