Endüstri Mühendisliği Bölümü Yayın Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/279
Browse
3 results
Search Results
Article Citation - WoS: 11Citation - Scopus: 10Ranking Using Promethee When Weights and Thresholds Are Imprecise: a Data Envelopment Analysis Approach(Taylor & Francis Ltd, 2022) Eryilmaz, Utkan; Karasakal, Orhan; Karasakal, EsraMulticriteria decision making (MCDM) provides tools for the decision makers (DM) to solve complex problems with multiple conflicting criteria. Scalarization of criteria values requires using weights for criteria. Determining weights creates controversy as they are influential on the final ranking and challenges the DM as they are hard to elicit. PROMETHEE method is widely used in MCDM for ranking the alternatives and appropriate in situations when there is limited information on the preference structure of the DM. The DM should provide exact values for parameters such as criteria weights and thresholds of preference functions. Data Envelopment Analysis (DEA) is used for measuring the relative efficiency of alternatives in a non-parametric way without requiring any weight input. In this study, we propose two novel PROMETHEE based ranking approaches that address the determination of weight and threshold values by using an approach inspired by DEA. The first approach can deal with imprecise specification of criteria weights, and the second approach can utilize both imprecise weights and thresholds. The proposed approaches provide the DM substantial flexibility on the required level of information on those parameters. An illustrative example and a real-life case study are presented to show the utility of the proposed approaches.Article Citation - WoS: 4Citation - Scopus: 4Inference in Multivariate Linear Regression Models With Elliptically Distributed Errors(Taylor & Francis Ltd, 2014) Yazici, Mehmet; Islam, M. Qamarul; Yildirim, FetihIn this study we investigate the problem of estimation and testing of hypotheses in multivariate linear regression models when the errors involved are assumed to be non-normally distributed. We consider the class of heavy-tailed distributions for this purpose. Although our method is applicable for any distribution in this class, we take the multivariate t-distribution for illustration. This distribution has applications in many fields of applied research such as Economics, Business, and Finance. For estimation purpose, we use the modified maximum likelihood method in order to get the so-called modified maximum likelihood estimates that are obtained in a closed form. We show that these estimates are substantially more efficient than least-square estimates. They are also found to be robust to reasonable deviations from the assumed distribution and also many data anomalies such as the presence of outliers in the sample, etc. We further provide test statistics for testing the relevant hypothesis regarding the regression coefficients.Article Citation - WoS: 20Citation - Scopus: 23Freight Transportation Using High-Speed Train Systems(Taylor & Francis Ltd, 2016) Ozcan, M. Keskin; Ertem, M. A.; Keskin Özcan, M.This study investigates the use of high-speed trains (HSTs) for transporting freight, such as small cargo and mail. A HST scheduling model is constructed to observe the effects of including freight in a passenger-only system. The proposed mathematical model is tested with an experimental study using the Turkish State Railways high-speed rail network and train sets. Freight transportation is analyzed in two cases, namely, adding separate freight trains to the system and using passenger trains for freight transportation. It can be concluded that dividing the sequences of cities into two allows for the completion of train services earlier in the day, and using the same train for transporting both passengers and freight provides more time saving in the system.