Endüstri Mühendisliği Bölümü Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.11779/1942
Browse
5 results
Search Results
Article Citation - WoS: 5Citation - Scopus: 12Predicting Cash Holdings Using Supervised Machine Learning Algorithms(Springer, 2022-05-18) Özlem, Şirin; Tan, Ömer FarukThis study predicts the cash holdings policy of Turkish firms, given the 20 selected features with machine learning algorithm methods. 211 listed firms in the Borsa Istanbul are analyzed over the period between 2006 and 2019. Multiple linear regression (MLR), k-nearest neighbors (KNN), support vector regression (SVR), decision trees (DT), extreme gradient boosting algorithm (XGBoost) and multi-layer neural networks (MLNN) are used for prediction. Results reveal that MLR, KNN, and SVR provide high root mean square error (RMSE) and low R2 values. Meanwhile, more complex algorithms, such as DT and especially XGBoost, derive higher accuracy with a 0.73 R2 value. Therefore, using advanced machine learning algorithms, we may predict cash holdings considerably.Article Citation - WoS: 54Citation - Scopus: 57Branch-And Methods for the Electric Vehicle Routing Problem With Time Windows(Taylor and Francis, 2021-07-31) Çatay, Bülent; Duman, Ece Naz; Taş, DuyguIn this paper, we address the electric vehicle routing problem with time windows and propose two branch-and-price-and-cut methods based on a column generation algorithm. One is an exact algorithm whereas the other is a heuristic method. The pricing sub-problem of the column generation method is solved using a label correcting algorithm. The algorithms are strengthened with the state-of-the-art acceleration techniques and a set of valid inequalities. The acceleration techniques include: (i) an intermediate column pool to prevent solving the pricing sub-problem at each iteration, (ii) a label correcting method employing the ng-route algorithm adopted to our problem, (iii) a bidirectional search mechanism in which both forward and backward labels are created, (iv) a procedure for dynamically eliminating arcs that connect customers to remote stations from the network during the path generation, (v) a bounding procedure providing early elimination of sub-optimal routes, and (vi) an integer programming model that generates upper bounds. Numerical experiments are conducted using a benchmark data set to compare the performances of the algorithms. The results favour the heuristic algorithm in terms of both the computational time and the number of instances solved. Moreover, the heuristic algorithm is shown to be specifically effective for larger instances. Both algorithms introduce a number of new solutions to the literature.Article Citation - WoS: 35Citation - Scopus: 36Large-Scale Influence Maximization Via Maximal Covering Location(Elsevier, 2021-02-01) Güney, Evren; Ruthmair, Mario; Sinnl, Markus; Leitner, MarkusInfluence maximization aims at identifying a limited set of key individuals in a (social) network which spreads information based on some propagation model and maximizes the number of individuals reached. We show that influence maximization based on the probabilistic independent cascade model can be modeled as a stochastic maximal covering location problem. A reformulation based on Benders decomposition is proposed and a relation between obtained Benders optimality cuts and submodular cuts for correspondingly defined subsets is established. We introduce preprocessing tests, which allow us to remove variables from the model and develop efficient algorithms for the separation of Benders cuts. Both aspects are shown to be crucial ingredients of the developed branch-and-cut algorithm since real-life social network instances may be very large. In a computational study, the considered variants of this branch-and-cut algorithm outperform the state-of-the-art approach for influence maximization by orders of magnitude.Article Citation - WoS: 16Citation - Scopus: 18Gradual Covering Location Problem With Multi-Type Facilities Considering Customer Preferences(Elsevier, 2020-09-01) Küçükaydın, Hande; Aras, NecatiIn this paper, we address a discrete facility location problem where a retailer aims at locating new facilities with possibly different characteristics. Customers visit the facilities based on their preferences which are represented as probabilities. These probabilities are determined in a novel way by using a fuzzy clustering algorithm. It is assumed that the sum of the probabilities with which customers at a given demand zone patronize different types of facilities is equal to one. However, among the same type of facilities they choose the closest facility, and the strength at which this facility covers the customer is based on two distances referred to as full coverage distance and gradual (partial) coverage distance. If the distance between the customer location and the closest facility is smaller (larger) than the full (partial) coverage distance, this customer is fully (not) covered, whereas for all distance values between full and partial coverage, the customer is partially covered. Both distance values depend on both the customer attributes and the type of the facility. Furthermore, facilities can only be opened if their revenue exceeds a certain threshold value. A final restriction is incorporated into the model by defining a minimum separation distance between the same facility types. This restriction is also extended to the case where a minimum threshold distance exists among facilities of different types. The objective of the retailer is to find the optimal locations and types of the new facilities in order to maximize its profit. Two versions of the problem are formulated using integer linear programming, which differ according to whether the minimum separation distance applies to the same facility type or different facility types. The resulting integer linear programming models are solved by three approaches: commercial solver CPLEX, heuristics based on Lagrangean relaxation, and local search implemented with 1-Add and 1-Swap moves. Apart from experimentally assessing the accuracy and the efficiency of the solution methods on a set of randomly generated test instances, we also carry out sensitivity analysis using a real-world problem instance.Article An Efficient Linear Programming Based Method for the Influence Maximization Problem in Social Networks (vol 503, Pg 589, 2019)(Elsevier, 2020-02-01) Güney, EvrenThe influence maximization problem (IMP) aims to determine the most influential individuals within a social network. In this study first we develop a binary integer program that approximates the original problem by Monte Carlo sampling. Next, to solve IMP efficiently, we propose a linear programming relaxation based method with a provable worst case bound that converges to the current state-of-the-art 1-1/e bound asymptotically. Experimental analysis indicate that the new method is superior to the state-of-the-art in terms of solution quality and this is one of the few studies that provides approximate optimal solutions for certain real life social networks.