Sürdürülebilir Kalkınma Amaçları
Yayınlar
- MakaleAçık Erişim2025Scopus SJRQ2Forecasting daily natural gas consumption with regression, time series and machine learning based methods
Yucesan, Melih; Pekel, Engin; Celik, Erkan; Gul, Muhammet; Serin, Faruk
An effective short-term natural gas forecasting method contributes to social contributions and allows industrial chain elements to function effectively and minimize economic losses. We dealt with a comparative framework on the applicability of different methods in daily natural gas service (NGS) consumption forecasting. In this context, time series, machine learning, evolutionary and population-based approaches, and their hybrid versions are applied to the NGS data. Hybridized approaches are tested in the scope of NGS consumption forecasting for the first time in the literature in this study. The case of Turkey is handled, and its NGS data is used to demonstrate the comparative framework’s applicability. The comparative study is assessed in the lights of common forecasting accuracy measures of mean absolute percentage error (MAPE), R-squared (R2), and mean squared error (MSE). According to each method’s results, the seasonal autoregressive integrated moving average with exogenous regressors (SARIMAX) and artificial neural network (ANN) hybrid model provides the most dominant performance with respect to MAPE. The lowest error was obtained with a MAPE value of 0.357 in this hybrid model constructed under seven neurons in its ANN structure. This model is followed by another hybrid model, autoregressive integrated moving average (ARIMA)-ANN, with a MAPE value of 0.5 under nine neurons in terms of accuracy performance. The worst performance value belongs to the Genetic algorithm-ANN hybrid model with a MAPE value of approximately 26%.
AtıfScopus16WoS19OpenAlex22
İncele - MakaleAçık Erişim2022Scopus SJRQ1How Covid-19 pandemic and partial lockdown decisions affect air quality of a city? The case of Istanbul, Turkey
Celik, Erkan; Gul, Muhammet
The world is currently struggling with a new type of coronavirus (2019-nCoV) pandemic that first appeared in Wuhan, China, and then spread to almost all countries. As in other countries of the world, public authorities in Turkey are implementing many preventive and mitigating partial lockdown (PL) actions against the virus's effects. Some decisions and policies implemented before and after March 11, 2020, when the first virus case has been identified, have reduced people and traffic circulation, which has also turned into some improvements in air quality. At this point, this study aims to investigate how this pandemic affects the air quality of a metropolis. A case study of the city of Istanbul, the most affected city with more than half of Turkey's cases, is performed. In our analysis, we observe, compare, and discuss the impact of the COVID-19 pandemic and PL decisions on Istanbul city's air quality. We consider the particulate matter (PM10), sulfur dioxide (SO2), carbon monoxide (CO), nitrogen dioxide (NO2), nitrogen oxide (NO), nitrogen oxides (NOx), and ozone (O3) concentrations. We used data from 19 air monitoring stations (AMSs) and obtained improvements in the air quality for the pandemic period. In summary, the concentration levels in PM10, NO2, NO, and NOx result in a clear decline in pandemic times compared to the normal times in Istanbul. On the other hand, a non-homogenous trend for SO2 and CO concentrations is observed for different AMSs. A partial increase in O2 concentration is obtained in the comparison of before and during the PL period.
AtıfScopus15WoS13OpenAlex22
İncele - Kitap BölümüYalnız künye2021Fine–kinney-based occupational risk assessment using interval-valued pythagorean fuzzy vikor
Gul, Muhammet; Mete, Suleyman; Serin, Faruk; Celik, Erkan
This chapter aims at the adaptation of the Fine–Kinney occupational risk assessment concept into the VIKOR multi-attribute decision-making method with an interval-valued Pythagorean fuzzy set. The classical fuzzy set theory has been improved by proposing a number of extended versions. One of them is the Pythagorean fuzzy set. It has been firstly developed by Yager (IEEE Transactions on Fuzzy Systems 22:958–965, [1]). In this chapter, we use this type of fuzzy set with VIKOR since it reflects uncertainty in occupational risk assessment and decision-making better than other fuzzy extensions. To demonstrate the proposed approach applicability, a case study regarding the activities of the surface treatment area in a chrome plating unit of a gun factory is performed. Some additional analysis to test the solidity and validity of the approach is executed. Finally, the Python codes in the implementation of the proposed approach are given for scholars and practitioners for usage in further studies.
AtıfScopus14OpenAlex5İncele - EditoryalYalnız künye2021Preface
Gul, Muhammet; Mete, Suleyman; Serin, Faruk; Celik, Erkan
AtıfScopus2İncele - Kitap BölümüYalnız künye2021Fine–kinney-based occupational risk assessment using interval type-2 fuzzy vikor
Gul, Muhammet; Mete, Suleyman; Serin, Faruk; Celik, Erkan
In this chapter, we improved Fine–Kinney occupational risk assessment approach with interval type-2 fuzzy VIKOR (IT2FVIKOR). VIKOR is a compromise multi-attribute decision-making method proposed by Opricovic (1998). Similar to other compromised solution-based approaches, it considers the solution which is closest to the ideal. In this chapter, we adapted the interval type-2 fuzzy sets (IT2FSs) into VIKOR as it reflects the uncertainty well in decision-making. IT2FVIKOR algorithm under the Fine–Kinney concept provides a useful and solid approach to the occupational health and safety risk assessment. In addition to proposing this new approach, a case study is performed in a gun and rifle barrel external surface oxidation and coloring unit of a gun factory. A validation and a sensitivity analysis is also attached to this study. Finally, the proposed approach is implemented in Python.
AtıfScopus4OpenAlex5İncele - MakaleAçık Erişim2021Scopus SJRQ2Metaheuristic Approaches Integrated with ANN in Forecasting Daily Emergency Department Visits
Pekel, Engin; Gul, Muhammet; Celik, Erkan; Yousefi, Samuel
The overall service quality level of Emergency Departments (EDs) can be improved by accurate forecasting of patient visits. Accordingly, this study aims to evaluate the use of three metaheuristic approaches integrated with Artificial Neural Network (ANN) in forecasting daily ED visits. To do this, five performance measures are used for evaluating the accuracy of the proposed approaches, including Bayesian ANN, Genetic Algorithm-based ANN (GA-ANN), and Particle Swarm Optimization algorithm-based ANN (PSO-ANN). The outputs of this study show that the PSO-ANN model provides the most dominant performance in both the training and testing process. The lowest error is obtained with a mean absolute percentage error (MAPE) of 6.3%, Mean Absolute Error (MAE) of 42.797, Mean Squared Error (MSE) of 2499.340, Root Mean Square Error (RMSE) of 49.933, and R-squared (R2) of 0.824 on the training dataset. The lowest error with an MAPE of 6.0%, MAE of 40.888, MSE of 2839.998, RMSE of 53.292, and R2 of 0.791 is also obtained on the testing process.
AtıfScopus7WoS5OpenAlex7İncele - Kitap BölümüYalnız künye2021Fine–kinney-based occupational risk assessment using fuzzy best and worst method (f-bwm) and fuzzy mairca
Gul, Muhammet; Mete, Suleyman; Serin, Faruk; Celik, Erkan
The best-worst method (BWM) proposed by Rezaei (Omega 53:49–57, [1]) is an MCDM method used to achieve the weights of the criteria by making fewer pairwise comparisons and using more consistent decision matrices. In this chapter, instead of crisp numbers, triangular fuzzy numbers that reflect the uncertainty well in real-world problems are used in integration with the BWM method (Guo and Zhao in Knowl Based Syst 121:23–31, [2]) in determining factor weights. Using the fuzzy best and worst method (F-BWM), a model based on Fine–Kinney occupational health and safety risk assessment method was developed for the first time in the literature. Three parameters of Fine–Kinney method are weighted by the mathematical models of F-BWM. Then, the risks are prioritized by fuzzy multi-attribute ideal real comparative analysis (F-MAIRCA). A case study was conducted to demonstrate the feasibility of the approach, and besides this case study, a comparative study was also conducted to test the validity of the proposed approach. This approach led to the conclusion that Fine–Kinney’s method, BWM, MAIRCA, and triangular fuzzy sets make the risk decision-making process more dynamic, taking into account the benefits of these methods individually or in integration.
AtıfScopus11OpenAlex12İncele - Kitap BölümüYalnız künye2021Fine–kinney-based occupational risk assessment using hexagonal fuzzy multimoora
Gul, Muhammet; Mete, Suleyman; Serin, Faruk; Celik, Erkan
Hexagonal fuzzy numbers (HFNs) can be used as a proficient logic to simplify understanding of ambiguity information. HFNs present the usual information in a comprehensive way and also the ambiguity section can be exemplified in a reasonable way. In this chapter, we proposed an improved Fine–Kinney occupational risk assessment approach using a well-known MCDM method Multi-Objective Optimization by Ratio Analysis (MULTIMOORA) using hexagonal fuzzy numbers. Since the mere MULTIMOORA has failed to handle uncertainty and vague information which usually exist in real world problems, we follow integration of HFNs and MULTIMOORA (HFMULTIMOORA). To show the applicability of the novel approach, a case study of risk assessment of a raw mill in cement plant was provided. Comparative analysis with using two aggregation tools as reciprocal rank method and dominance theory are carried out. Finally, the Python implementation of the proposed approach is implemented to be effective for those concerned in the future.
AtıfScopus7OpenAlex5İncele - Kitap BölümüYalnız künye2021Fine–kinney-based occupational risk assessment using intuitionistic fuzzy todim
Gul, Muhammet; Mete, Suleyman; Serin, Faruk; Celik, Erkan
This chapter applies a novel occupational risk assessment approach which merges TODIM with the Fine–Kinney method under the intuitionistic fuzzy set concept. Risk parameters of Fine–Kinney and OHS experts are weighted by an intuitionistic fuzzy weighted averaging (IFWA) aggregation operator. Hence, hazards are quantitatively evaluated and prioritized using the proposed approach. To illustrate the novel risk assessment approach, processes of the gun and rifle assembly line of a factory are handled. A comprehensive risk assessment is carried out to improve operational safety and reliability in the industry. We adapt intuitionistic fuzzy sets in the existing study since they reflect uncertainty with the aid of their membership and nonmembership functions in decision-making better than classical fuzzy extensions. An additional sensitivity analysis by changing the attenuation parameter of TODIM is performed to test the validity of the approach. Finally, the Python codes in the implementation of the proposed approach are given for scholars and practitioners for usage in further studies.
AtıfScopus5OpenAlex9İncele - Kitap BölümüYalnız künye2021Fine–kinney-based occupational risk assessment using single-valued neutrosophic topsis
Gul, Muhammet; Mete, Suleyman; Serin, Faruk; Celik, Erkan
Neutrosophic sets are initially recommended by Smarandache (First international conference on neutrosophy, neutrosophic logic, set, probability, and statistics. University of New Mexico, Gallup, NM, pp 338–353, 2002 [1]). These sets reflect uncertainty and vagueness in real-world problems better than classical fuzzy set theory. It takes into consideration three decision-making situations called indeterminacy, truthiness, and falsity. In Zadeh traditional fuzzy set theory, there is just membership function fuzzy set degree. But, in neutrosophic environment, it considers three membership functions. Unlike intuitionistic fuzzy sets, an indeterminacy degree is considered. In this chapter, we applied a special form of neutrosophic set as single-valued neutrosophic set (SVNs) with the technique for order preference by similarity to ideal solution (TOPSIS) under the concept of Fine–Kinney occupational risk assessment. Since the mere TOPSIS has failed to handle imprecise and vague information which usually exist in real-world problems, we follow the integration of SVNs and TOPSIS. To demonstrate the applicability of the novel approach, a case study of risk assessment of a wind turbine in times of operation was provided. Comparative analysis with some similar approaches and sensitivity analysis by changing the weights of Fine–Kinney parameters are carried out. Finally, the Python implementation of the proposed approach is executed to be useful for those concerned in the future.
AtıfScopus12OpenAlex9İncele - MakaleYalnız künye2021Scopus SJRQ1Hazard identification, risk assessment and control for dam construction safety using an integrated BWM and MARCOS approach under interval type-2 fuzzy sets environment
Celik, Erkan; Gul, Muhammet
Undoubtedly, dams are one of the projects developed by people to make optimal water use and generate hydroelectric energy power. In addition to the importance of detailed project planning, faultless production, and optimal working conditions in dam constructions, it is equally important to identify and analyze the risks from the occupational health and safety (OHS) perspective that may arise during construction and operation. In this context, hazard identification, risk analysis, and control, which are the main issues of dam safety, have gained importance. In this study, a novel approach is developed for dam construction safety. Two newly proposed multi-criteria decision-making (MCDM) methods of “Best-Worst Method-BWM” and “measurement of alternatives and ranking to Compromise solution-MARCOS”, are merged under the context of interval type-2 fuzzy sets (IT2FSs). Severity and probability, two essential parameters of the risk score, have been weighted using BWM. Subsequently, the priority orders of hazards have been determined by the MARCOS. In implementing the approach to the blasting process in dam construction, several control measures are suggested for stakeholders. Some comparative works and sensitivity analyses are performed to test the approach's validity and solidity from the methodological viewpoint.
AtıfScopus140WoS113İncele - MakaleAçık Erişim2021A holistic FMEA approach by fuzzy-based Bayesian network and best–worst method
Yucesan, Melih; Gul, Muhammet; Celik, Erkan
Failure mode and effect analysis (FMEA) is a risk analysis tool widely used in the manufacturing industry. However, traditional FMEA has limitations such as the inability to deal with uncertain failure data including subjective evaluations of experts, the absence of weight values of risk parameters, and not considering the conditionality between failure events. In this paper, we propose a holistic FMEA to overcome these limitations. The proposed approach uses the fuzzy best–worst (FBWM) method in weighting three risk parameters of FMEA, which are severity (S), occurrence (O), and detection (D), and to find the preference values of the failure modes according to parameters S and D. On the other side, it uses the fuzzy Bayesian network (FBN) to determine occurrence probabilities of the failure modes. Experts use a procedure using linguistic variables whose corresponding values are expressed in trapezoidal fuzzy numbers, and determine the preference values of the failure modes according to parameter O in the constructed BN. Thus, the FBN including expert judgments and fuzzy set theory addresses uncertainty in failure data and includes a robust probabilistic risk analysis logic to capture the dependence between failure events. As a demonstration of the approach, a case study was conducted in an industrial kitchen equipment manufacturing facility. The results of the approach have also been compared with existed methods demonstrating its robustness.
AtıfScopus105WoS93İncele - MakaleYalnız künye2021Scopus SJRQ1Green supplier selection for textile industry: a case study using BWM-TODIM integration under interval type-2 fuzzy sets
Celik, Erkan; Yucesan, Melih; Gul, Muhammet
Although environmental awareness has reached a high level, enterprises—regardless of their working domains—follow the concept of greenness for their practices. This awareness among the stakeholders and supply chain experts has a positive impact on the purchasing departments of enterprises in various sectors to consider greenness in their procurement processes. The critical decision that must be made in green supply chain management (GSCM) is supplier selection. In the textile industry, a highly competitive market in recent years, suppliers for this industry have crucial roles in business activities considering environmental issues. Therefore, green supplier selection (GSS) in the textile industry is considered a must-be process for the stakeholders. In this study, a GSS problem is tackled as a multi-criteria decision process. Best worst method (BWM) and TODIM (an acronym in Portuguese of interactive and multi-criteria decision-making) methods are merged under an improved fuzzy concept of interval type-2 fuzzy sets (IT2FSs). In determining green suppliers’ evaluation criteria, BWM with interval type-2 fuzzy numbers (IT2F-BWM) is used. In selecting green suppliers, an interval type-2 fuzzy TODIM (IT2F-TODIM) is applied. Considering the characteristics of IT2FSs, BWM, and TODIM methods either individually and in integrated style, the proposed approach can handle uncertainty in the decision-making of GSS. To demonstrate the applicability of the approach, a case study in the Turkish textile industry is performed. Three green supplier alternatives (S1, S2, and S3) are assessed under forty-two sub-criteria. The study shows the most significant sub-criteria are recognized as dye and print quality, product design and pattern suitability, profit on the product, variation in price, and purchase cost. S2 green supplier has been selected as the most appropriate one. A sensitivity analysis is also fulfilled to check variation in the ranking of green suppliers.
AtıfScopus88WoS72OpenAlex100İncele - Kitap BölümüYalnız künye2021Fine–kinney occupational risk assessment method and its extensions by fuzzy sets: A state-of-the-art review
Gul, Muhammet; Mete, Suleyman; Serin, Faruk; Celik, Erkan
The Fine–Kinney method (Fine in J Saf Res 3:157–166, 1971; Kinney and Wiruth in Practical risk analysis for safety management. Naval Weapons Center, pp 1–20, 1976), which was first introduced as an occupational health and safety risk analysis tool in the 1970s, is a systematic methodology that provides a mathematical formula for calculating the risk that arises due to a specified hazard. In the traditional version of Fine–Kinney as suggested in its original version, a risk score (RS) is calculated as a result of mathematical multiplication of probability (P), exposure (E), and consequence (C) parameters. These calculated risk scores are used to establish priorities for the corrective efforts in order to eliminate risks or reduce their effects to a reasonable level. This simple and useful method is preferred and implemented by small and medium-sized enterprises. In the academic literature, it has been applied for many risk analysis problems, although it includes several drawbacks recently revealed. In this method, no weight assignment is made for each risk parameter. Also, it is hard to assess consequence, exposure, and probability, precisely. Multi-criteria decision making (MCDM) is a pool of methods used in occupational health and safety risk analysis both by international standard-setting organizations and scholars from the literature. In classical MCDM methods, performance values and weights of decision criteria are known precisely and are specified with crisp numbers. However, many real-world problems contain uncertainties, and the knowledge and judgment of experts cannot be expressed precisely. Fuzzy-based MCDM methods, which are developed to reflect types and degrees of uncertainties better, produce more accurate results compared to classical methods. In this chapter, we first present the basics of Fine–Kinney method, including its implementing procedure, basic terminology, and drawbacks. Then, we provide a state-of-the-art review of Fine–Kinney occupational risk assessment method and its extensions by fuzzy sets. Graphical results obtained from the review are demonstrated to show the current state. Future work suggestions are also included to the chapter to show the possible gaps and possible opportunities.
AtıfScopus12OpenAlex12İncele - Kitap BölümüYalnız künye2021Fine–kinney-based occupational risk assessment using interval type-2 fuzzy qualiflex
Gul, Muhammet; Mete, Suleyman; Serin, Faruk; Celik, Erkan
In this chapter, we improved Fine–Kinney occupational risk assessment approach with interval type-2 fuzzy QUALIFLEX (IT2FQUALIFLEX). QUALIFLEX is an outranking multi-attribute decision-making method proposed by an extension of the Paelinck’s (Pap Reg Sci Assoc 36:59–74,[1]), generalized Jacquet-Lagreze’s permutation method. Similar to other outranking solution-based approaches, it considers the solution which is a comparison of hazards. In this chapter, we adapted the interval type-2 fuzzy sets (IT2FSs) into QUALIFLEX as it reflects the uncertainty well in decision-making. IT2FQUALIFLEX algorithm under the Fine–Kinney concept provides a useful and solid approach to the occupational health and safety risk assessment. In addition to proposing this new approach, a case study is performed in the chrome plating unit. A validation is also performed in this study. Finally, the proposed approach is implemented in Python.
AtıfScopus7OpenAlex7İncele - Kitap BölümüYalnız künye2021Fine–kinney-based occupational risk assessment using interval type-2 fuzzy topsis
Gul, Muhammet; Mete, Suleyman; Serin, Faruk; Celik, Erkan
This chapter proposed an improved Fine–Kinney occupational risk assessment approach using a well-known MCDM method “TOPSIS” under interval type-2 fuzzy set concept. It is defined as a technique for order preference by similarity to ideal solution by Hwang and Yoon [1]. It is based on separation from ideal and anti-ideal solution concept. Since the initial crisp data-based version is insufficient by time in reflecting the uncertainty in decision-maker’s opinions, fuzzy sets are integrated to the TOPSIS algorithm to provide a solid and comprehensive method. Interval type-2 fuzzy set is an improved version of the type-1 fuzzy set. It is also a special version of a general type-2 fuzzy set. Since general type-2 fuzzy systems contain complex computational operations, they cannot be easily applied to real-world problems such as occupational risk assessment. Interval type-2 fuzzy sets are the most frequently used type-2 fuzzy sets due to their ability in handling more uncertainty and producing more accurate and solid results. The Fine–Kinney concept is merged with the interval type-2 fuzzy set concept and TOPSIS for the first time through the literature. To demonstrate the applicability of the proposed approach, a case study is carried out in a chrome plating unit of a gun factory. Some beneficial validation and sensitivity analysis are also performed. Finally, as a creative contribution of our book, the implementation of the proposed approach in Python is performed.
AtıfScopus2OpenAlex3İncele - KitapYalnız künye2020Computational intelligence and soft computing applications in healthcare management science
Gul, Muhammet; Celik, Erkan; Mete, Suleyman; Serin, Faruk
In today's modernized world, the field of healthcare has seen significant practical innovations with the implementation of computational intelligence approaches and soft computing methods. These two concepts present various solutions to complex scientific problems and imperfect data issues. This has made both very popular in the medical profession. There are still various areas to be studied and improved by these two schemes as healthcare practices continue to develop.
AtıfOpenAlex7İncele - EditoryalYalnız künye2020Preface
Gul, Muhammet; Celik, Erkan; Mete, Suleyman; Serin, Faruk
AtıfScopus1İncele - MakaleYalnız künye2020Scopus SJRQ1A manufacturing failure mode and effect analysis based on fuzzy and probabilistic risk analysis
Gul, Muhammet; Yucesan, Melih; Celik, Erkan
This study proposes an improved failure mode and effect analysis (FMEA) with fuzzy Bayesian Network (FBN) and fuzzy best-worst method (FBWM) to assess failures in plastic production. To eliminate the drawbacks of the classical risk priority number (RPN) computation of FMEA, this approach is developed. Unlike classical RPN, a new parameter hierarchy is constructed, and three sub-parameters are injected into the approach. These parameters are weighted by the aid of FBWM. Hereafter, a fuzzy rule-based system is constructed by incorporating Bayesian Network (BN). Also, a sensitivity analysis is performed to observe the final FMEA score changes in accordance with the change of subjective probability values. Finally, a comparative analysis with two approaches of classical FMEA and FBWM-based FMEA (without a fuzzy rule-based system incorporating BN) is fulfilled. The results of the study are strengthened with the experts’ opinions regarding the importance of failure modes for the final product and the whole system and supported them by experience feedback in the observed facility.
AtıfScopus68WoS59OpenAlex68İncele - Kitap BölümüYalnız künye2020NARX neural networks model for forecasting daily patient arrivals in the Emergency Department
Yucesan, Melih; Mete, Suleyman; Serin, Faruk; Celik, Erkan; Gul, Muhammet
Regarding measuring of service quality at the emergency departments (ED), essential parameters are length of stay (LOS) and waiting times. Patient arrivals, which is related to LOS and waiting times, is hard to forecast and is affected by many parameters. Therefore, authors employed a Nonlinear Autoregressive Exogenous (NARX) model for forecasting of ED arrivals. NARX models are used extensively in many applications that show non-linear and dynamic behavior, but as far as authors know, the NARX method has not yet been used in the forecast of ED arrivals before. In this study, calendar and climatic variables are defined as input parameters. Patient Arrivals is defined as output parameter. A commercial software, MATLAB, was used to train and test the data set. To find the best network architecture Levenberg- Marquardt (LM) and Bayesian Regularization (BR) algorithms, different lags, and number of neurons were tested. R-squared and mean square error (MSE) are used to evaluate the accuracy of the tested networks.
AtıfScopus2OpenAlex4İncele