Mühendislik Fakültesi / Faculty of Engineering

Permanent URI for this collectionhttps://hdl.handle.net/11727/1401

Browse

Search Results

Now showing 1 - 3 of 3
  • Item
    Evaluation of Effect of Different Membership Functions on Risk Assessment
    (2018) Atalay, Kumru Didem; Can, Guelin Feryal; Eraslan, Ergun; 28330411
    This study aims to define the relationship between risk degrees and risk indexes on different functional structures with the assumption that risk degrees may not always present a linear relationship with the risk indexes. In this way, risk indexes suitable for expert evaluation of working conditions and computed using three different membership functions are determined. Among the membership functions used, one is preferred as linear and the others are preferred as non-linear. Additionally, a new fuzzy risk assessment (RA) algorithm is developed using these three membership functions. With this new fuzzy RA algorithm, a more flexible and precise process becomes available, while information loss during the determination of the risk index of danger sources is prevented. As a result, non-linear increasing membership function is selected as most suitable for the expression of the relationship between risk degrees and risk indexes.
  • Item
    An improved decision support system for ABC inventory classification
    (2020) Eraslan, Ergun; Ic, Yusuf Tansel
    In this study, an Improved Decision Support System (IDSS) is developed to help the decision makers in their inventory classification decisions. For the first time in this paper, the novel IDSS for ABC classification is developed. Certain new algorithms regarding the manufacturing company's features are applied in a framework of IDSS. The IDSS is developed as modular structure and provided the integrated modules of "Data-Base" and "ABC Analysis". In the developed IDSS, the appropriate ABC classification models are considered among Annual Dollar Usage (ADU), Analytic Hierarchy Process (AHP), Scoring (SCR), Fuzzy C-means Algorithm (FCM), and Analytic Network Process (ANP). Some issues and applicability of the IDSS are illustrated with real case problems in the paper. The proposed IDSS software is considerably decreased the time for the inventory classification. In the meantime it could be easily used in various sectors. Therefore, the proposed IDSS significantly contributed to obtaining more accurate and quickly modifiable ABC classification in real cases. Furthermore, the user friendly software can be updated readily according to recent developments in the market.
  • Item
    Development of a Spreadsheet DSS for Multi-Response Taguchi Parameter Optimization Problems Using the TOPSIS, VIKOR, and GRA Methods
    (2019) Kececi, Barış; Ic, Yusuf Tansel; Eraslan, Ergun; 0000-0002-2730-5993; AAI-1081-2020
    This paper presents a spreadsheet-based decision support system (DSS) for any parameter optimization problem, in the small- and medium-sized enterprises to help the managers to make better decisions. Microsoft Excel is used as a DSS development platform. The DSS application requires the quality characteristics and the level of parameters affecting the problem. The proposed system considers three multi-criteria decision-making methods: TOPSIS, VIKOR and GRA. These methods are integrated into the Taguchi method to convert the multi-response optimization problem to a single-response problem. The DSS suggests proper Taguchi experimental designs and provides the decision maker with an opportunity to use different metrics and to validate the experimental results. Several issues and an application are provided for illustrative purposes. The proposed DSS is tested on a case study (the performance of the mixed integer programming (MIP) formulation solver) and the results highlight that the system is capable of offering satisfactory outcomes. Using such a quick and flexible DSS might help to reduce the daily workload of the decision makers. The different metrics used for the response variables which results with the different parameter combination. Using the optimal parameter combination of TOPSIS (come to the fore in case MinBest metric used), the MIP formulation solver gives the best integer objective function value of 609 and a GAP value of 1.93%, both of which are less than the values obtained using the other methods. Using the optimal parameter combination of GRA (come to the fore in case OptBest metric used), the MIP formulation gives a best integer objective function value of 632 and a GAP value of 6.52%, both of which are less than the values obtained by using the other methods.