Саобраћајни факултет [Научни радови] / Faculty of Transport and Traffic Engineering [Scientific papers]
Трајни URI за ову колекцију
Потражи
Недавне прослеђене ставке
Сада се приказује 1 - 5 од 150
- СтавкаAnalysis of Software Effort Estimation by Machine Learning Techniques(LIETA, 2023) Meharunnisa; Saqlain, Muhammad; Abid, Muhammad; Awais, Muhammad; Stević, ŽeljkoSoftware effort estimation is a crucial activity in software project management that involves predicting the level of effort required to develop or maintain software applications. Accurate estimates enable effective planning and staffing which are key to on-time and on-budget delivery of software projects. This paper presents an analysis of using machine learning techniques for improving software effort estimation based on empirical datasets. Five public datasets from various sources were used - ISBSG, NASA93, COCOMO, Maxwell, and Desharnais. The data was preprocessed by handling missing values, converting categorical features, and splitting into train-test sets. Four machine learning regression algorithms were evaluated-linear regression, Gradient Boosting, Random Forest, and Decision Tree. Additionally, correlation-based feature selection was applied to select relevant subset of features and reduce dimensionality. The comparative analysis focused on two key metrics -R2 and root mean squared error (RMSE) to evaluate prediction accuracy. The results indicate that linear regression and Random Forest models perform significantly better than other approaches for this effort estimation task when using correlation to select features. The best R2 scores were achieved for NASA93, COCOMO, Maxwell, and Desharnais datasets. RMSE was lowest for the Desharnais dataset indicating high accuracy. The findings suggest that correlation- based feature selection can improve machine learning models for software effort estimation. The strengths of linear regression and Random Forest models make them suitable for developing reliable estimation tools. The insights from this comparative analysis establish a strong baseline for future research. Software project planners can leverage these findings to build intelligent data-driven effort prediction systems
- СтавкаSustainable development solutions of public transportation:An integrated IMF SWARA and Fuzzy Bonferroni operator(Elsevier, 2023) Moslem, Sarbast; Stević, Željko; Tanackov, Ilija; Pilla, FrancescoDeveloping the quality of public transport has an efficient impact to attract more users to switch modes from private vehicles to public transport, which has a tremendous capability in reducing the traffic congestion, noise and CO2 emissions the urban areas. For this reason, policy makers and scholars aim to enhance the supply quality of public transport system, where involving citizens along with experts and decision makers in the decision process will reflect the existed and future demand to provide more sustainable solutions which provide efficient solutions to achieve positive impacts on the local environment. This study intends to determine the preference of decision-makers and operators for the importance of the supply quality elements of urban bus transport services in Mersin city, Turkey. For this aim, decision makers and experts are involved in the evaluation process to provide feedback on public transport quality with the view of increasing the satisfaction and thus usage, with positive impacts on increased public transport modal share and dropped CO2 emission transport. In total we have formed list of three main criteria: transport quality, service quality and traceability which contains in total 21 sub-criteria. For determining all criteria weights we have formed in total 112 models using IMF SWARA (Stepwise Weight Assessment Ratio Analysis) method and for final determining criteria weights Fuzzy Bonferroni operator (FBO) has been used. Applying such model we have ensured stability in final values of criteria and have obtained optimal results based on preferences decision makers. The adopted results show that the most significant attribute in the system is the “Traceability” with highest weight score (0.398), followed by the “Transport quality” with weight score (0.334), however, the service quality rank as the last significant attribute with weight score (0.264). The originality of our work is conducting IMF SWARA method for improving the service quality of the public transport system.
- СтавкаA NEW METHODOLOGY FOR TREATING PROBLEMS IN THE FIELD OF TRAFFIC SAFETY: CASE STUDY OF LIBYAN CITIES(Vilnius Gediminas Technical University, 2023) BADI, Ibrahim; STEVIĆ, Željko; RADOVIĆ, Dunja; RISTIĆ, Bojana; CAKIĆ, Aleksandar; SREMAC, SinišaTraffic safety is an area of great importance, since there are many traffic accidents every day in which a significant number of people are killed. Defining certain strategies and identifying potentially the most dangerous towns and cities regarding this area are, on the one hand, a necessity, and, on the other hand, a challenge. In this paper, integrated Multi-Criteria Decision-Making (MCDM) model for ranking cities in Libya from the aspect of traffic safety has been proposed. The model implies a set of 8 criteria on the basis of which 5 decision-makers rated the 10 most deprived cities in Libya. The Full Consistency Model (FUCOM) in combination with the rough Dombi aggregator is used to determine the significance of the criteria. The Rough Simple Additive Weighting (R-SAW) method is used to rank the alternatives. The rough Dombi aggregator is also used for averaging in group decision-making while evaluating the alternatives. The stability of the model and the obtained results has been verified by the sensitivity analysis, which implies a 2-phase procedure. In the 1st phase, rough Additive Ratio Assessment (R-ARAS), Rough Weighted Aggregated Sum Product Assessment (R-WASPAS), Rough Complex Proportional Assessment (R-COPRAS) and Rough Multi-Attributive Border Approximation-area Comparison (R-MABAC) methods are applied. The 2nd phase implies changing the parameter ρ in the procedure of rough Dombi aggregator, while the 3rd phase includes the calculation of Spearman’s Correlation Coefficient (SCC) that shows a high correlation of ranks.
- СтавкаAssessment ofMountain Tourism Sustainability Using Integrated Fuzzy MCDM Model(MDPI, 2023) Xu, Ming; Bai, Chunjing; Shi, Lei; Puška, Adis; Štilic, Anđelka; Stević, ŽeljkoThe sustainable development of mountain tourism is crucial for preserving the delicate ecosystems and resources found in these unique landscapes. This research paper investigates the sustainability of mountain lodges, which serve as essential facilities for delivering mountain tourism services. To assess sustainability, expert decision making involving eight selected experts was employed. A hybrid approach combining the IMF SWARA (IMproved Fuzzy Step-wise Weight Assessment Ratio Analysis) method with Fuzzy Dombi Aggregation Operators was utilized to determine the weights of various sustainability criteria. The IMF SWARA method assigned initial weights based on expert input, which were subsequently adjusted using Fuzzy Dombi Aggregation Operators. The findings highlight the significance of two key criteria as per expert evaluations: the quality of the services offered (C21) and the preservation of natural resources (C15). To rank and evaluate the mountain lodges, the fuzzy CRADIS (Compromise Ranking of Alternatives from Distance to Ideal Solution) method was employed, ultimately identifying Zabrana (ML6) as the top-ranked mountain lodge. The validity of these results was confirmed through result validation and sensitivity analysis. This research contributes by providing insights into the current state of mountain tourism and offering guidelines for enhancing the overall mountain tourism experience through the integration of fuzzy methods.
- СтавкаMCDM MODEL FOR CRITICAL SELECTION OF BUILDING AND INSULATION MATERIALS FOR OPTIMISING ENERGY USAGE AND ENVIRONMENTAL EFFECT IN PRODUCTION FOCUS(Vilnius Gediminas Technical University, 2023) ULUTAŞ, Alptekin; BALO, Figen; MIRKOVIĆ, Katarina; STEVIĆ, Željko; MOSTAFA, Mohamed M. H.In the context of sustainable buildings, an ecological study of building and insulating materials is critical since it may assist affirm or shift the path of new technology development. Utilising sustainable material is a part of the sustainable improvement. For this reason, material fabrication is the primary process for the energy usage and release of intense environmental gaseous. The fabrication of the insulation and building materials, as in every fabrication process, comprises an energy consumption of crude materials in addition to the pollutants’ release. In buildings, insulation is a relevant technological resolution for cutting energy usage. This study aims to assess the primary energy consumption and the environmental effects of the fabrication of building and thermal isolation materials by using a new hybrid MCDM model. The proposed new hybrid MCDM model includes Fuzzy FUCOM, CCSD and CRADIS methods. While the subjective weights of the criteria were determined with the fuzzy FUCOM method, the objective weights of the criteria were determined with the CCSD method. Construction materials were listed with the CRADIS method. According to the fuzzy FUCOM method, the most important criterion was determined as the CR3 criterion, while the most important criterion according to the CCSD method was determined as the CR1 criterion. According to the combined weights, the most important criterion was determined as the CR3 criterion. According to the CRADIS method, the material with the best performance was determined as Cement Plaster. The methodology used in this study is a novel approach therefore it has not been used in any study before. In addition, since the CRADIS method is a newly developed MCDM method, the number of articles related to this method is low. Therefore, this research gap will be filled with this study.