Instance-based comparative assessment with application in manufacturing

6th ModTech International Conference - Modern Technologies in Industrial Engineering

Afteni, C.^1 ; Frumuanu, G.^1 ; Epureanu, A.^1

Dunrea de Jos University, Manufacturing Engineering Department, Domneasc Street 111, Galai

800201, Romania^1

关键词: Comparative assessment;    Consumed energy;    Distance functions;    Jobs scheduling;    Manufacturing activities;    Multiple nonlinear regression;    Order acceptance;    Selection decisions;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/400/4/042001/pdf
DOI  :  10.1088/1757-899X/400/4/042001

来源: IOP

【 摘 要 】

The management of MTO-based production involves multiple decisions needing to be taken. They concern all stages of manufacturing activity - order acceptance, products design, processes planning and jobs scheduling. Such decision consists in selecting, at a given moment, the most suitable alternative of the potential ones. This means that comparative assessment of potential alternatives is required. The usual method underpinning the comparative assessment supposes the direct, separate evaluation of each alternative, before decision making. In this paper we suggest a different approach in performing the comparative assessment, based on alternatives rankings, considered as the most relevant information about them. More specific, this means that rankings are assigned to potential alternatives, by referring them to the cases of already performed manufacturing activities, recorded as past instances database, after ranking criteria such as cost, time span, consumed energy etc. Thus, the selection decision results by comparing potential alternatives rankings. For finding the ranking of a given alternative, a solution to assess the difference between instances is needed, at first. Then, after iteratively defining instances neighborhoods from database and modeling them by multiple nonlinear regression, its ranking is determined. Here, we propose an expression for the distance-function together with an algorithm for actually finding the ranking of the analyzed alternative. A numerical simulation for the instance-based comparative assessment, with the help of an instances artificial database is also presented.