Supplier evaluation and selection is a kind of problem which includes multiple criteria of qualitative and quantitative properties. From different alternatives it requires to find the best option using different criteria and opinions of the decision makers. Because of the judgments or the bias of the decision makers, sometimes classical methods cannot be precise. In this paper, a relatively new decision method called TODIM (an acronym in Portuguese for iterative multi-criteria decision-making-"Tomada de Decisão Iterativa Multicritério") is improved with fuzziness to prevent the above mentioned problems of the classical methods. A real life case study for a furniture manufacturing company is also be solved.

Purpose: The purpose of this paper is to provide a structured methodology to permit the optimal selection of the best‐suited computerized maintenance management system (CMMS) software within maintenance information technologies.Design/methodology/approach: The analysis has been executed adopting a multi‐attribute decision‐making methodology, namely the technique for order preference by similarity to an ideal solution (TOPSIS). For the selection process, 17 criteria under five main heading have been defined. Data obtained from questionnaires and interviews with the company's maintenance managers have been used in fuzzy TOPSIS.Findings: The application of the proposed approach allows the maintenance practitioners to concentrate on a limited subset of CMMS applications and to compare their actual capabilities in order to select the right one, rather than considering only their purchase cost.Research limitations/implications: Comparisons with other multi‐attribute decision‐making techniques such as AHP (analytic hierarchy process) and ELECTRE (elimination and choice expressing reality) under fuzzy conditions can be done for further research.Practical implications: This paper is a very useful source of information both for maintenance managers and stakeholders in making decisions about the selection of CMMS software.Originality/value: This paper addresses CMMS software evaluation and selection criteria for practitioners and proposes a new multi‐attribute decision‐making methodology, hierarchical fuzzy TOPSIS, for the problem.

Bu çalışma, Borsa İstanbul Anonim Şirketi (BİST) 30 Endeksi’nde işlem gören firmaların hisse senetlerinin gelecek fiyatlarını tahmin etmeyi amaçlamaktadır. Bu amaçla öncelikle BİST 30 Endeksi firmalarının 2010-2019 yılları arasındaki üçer aylık finansal tabloları temin edilmiş daha sonra bu tablolar vasıtasıyla firmalara ait finansal oranlar hesaplanmıştır. Ayrıca firma hisse senetlerinin aylık kapanış fiyatlarına ulaşılmış ve firmalara ait finansal oranlarla denk olacak şekilde üçer aylık ortalamaları alınmıştır. Bu şekilde veriler temin edildikten sonra Yapay Sinir Ağları (YSA), Rastgele Orman (RO) algoritması ve XGBoost algoritması kullanılarak her bir firmaya ait hisse senedinin gelecek fiyatı tahmin edilmiştir. Daha sonra her bir yönteme göre elde edilen tahmin sonuçları karşılaştırılmıştır. XGBoost ve Rastgele Orman algoritmaları birbirlerine yakın sonuçlar vermelerine rağmen XGBoost algoritması en iyi sonucu vermektedir. Ayrıca her iki modelin de YSA’ya göre daha yüksek performans gösterdiği tespit edilmiştir.