Scientia Iranica
Volume:25 Issue: 2, Mar - Apr 2018

Public transport crew scheduling is a worldwide problem, which is NP-hard. This paper presents a new crew scheduling approach, called GRAVIG, which integrates grey relational analysis (GRA) into a Variable Iterated Greedy (VIG) algorithm. The GRA is served as a solver for the shift selection during the schedule construction process, which can be considered as a multiple attribute decision making (MADM) problem, since there are multiple static and dynamic criteria governing the efficiency of a shift to be selected into a schedule. Moreover, in the GRAVIG, a biased probability destruction strategy is elaborately devised to keep the ‘good’ shifts remained in the schedule without compromising the randomness. Experiments on eleven real-world crew scheduling problems show that the GRAVIG can generate high-quality solutions close to the lower bounds obtained by the CPLEX in terms of the number of shifts.

Due to the increase in investments in construction projects and the lack of practical models in this area developing new practical models is essential. In this paper, researchers suggest a new model in which (1) its assumptions are adopted based on the real world, (2) goal programming is used because of the soft nature of the budget constraints; and (3) risk of variations in cash flows is considered. The presented model chooses the most profitable portfolio of projects and determines their respective financing resources, area under construction, and pre-sale and sale amounts for each period such that the cumulative cash flow at the end of the time horizon is maximized. The fuzzy analytic hierarchy process (FAHP) is used to determine the weight of the objectives. The exact solution to the model is obtained using the ILOG CPLEX software. The presented solution seems efficient; since it yields very small elapsed times to exactly solve the real-world-sized problems. Also, the sensitivity analysis is performed and the results are deliberately studied and analyzed. Parameters such as pre-sale prices, mean and variance of the sale price and construction costs are among the highly sensitive parameters.

In this study, an optimal integrated vendor-buyer inventory model with defective items is proposed. Most researches for defective items assumed that an inspection process is carried out by the buyer. We consider that the vendor conducts the inspection process and disposes defective items in multiple batches. We prove that the function of annual cost is convex, and obtain closed-form expressions. A solution procedure is used to derive the optimal order quantity, the number of shipments and the number of defective item disposals. Numerical examples are provided to illustrate our model. Setting the fraction of defective items be zero, the numerical examples indicate that the proposed model can result in the solutions of the existing models without considering defective items. Moreover, a sensitivity analysis is used to reveal the effects of cost parameters on the optimal solution. When the disposal cost is relatively low, we show that a multiple disposals strategy may perform better than a single disposal strategy.

The existing relational network data envelopment analysis (DEA) models evaluate the performance of decision making units (DMUs) with precise data. Whereas in the real world applications, there are many supply chain (SC) networks with imprecise and vague figures. This paper develops a relational network DEA model for evaluating the performance of supply chains with fuzzy numbers. The proposed fuzzy model is capable of evaluating the performance of all kinds of network structures. A pair of two-level mathematical program is utilized to convert the fuzzy relational network DEA to a conventional crisp one. For this purpose, the upper and lower bounds of the efficiencies are calculated by α-cut concept. The proposed model is implemented using actual data from the supply chain of an international shipping company in Iran.

This paper studies a bi-level decentralized supply chain consisting of one manufacturer and one distributor. Along with global advertising, the manufacturer participates in part of local advertising expenditure incurred by the distributor. Bi-level programming is applied to model the relationship between the manufacturer and distributer under two power scenarios of Stackelberg game framework. In the first scenario, we consider the manufacturer as the leader and in the latter, we allow the distributer to act as the dominant member of the supply chain. In order to tackle bi-level programming models, a meta-heuristic genetic algorithm with hierarchical structure is developed for each scenario and optimal policies for the members in terms of pricing, advertising, inventory and demand allocation are determined. Finally, several computational experiments are conducted on data obtained from an automotive spare parts supply chain to demonstrate the validity of the developed models and compare the benefits of members as well as of the entire system.

In traditional scheduling problems and in many real-world applications the production operations are scheduled regardless of distribution decisions. Indeed, the completion time of a job in such problems is traditionally defined as the time when the production sequences of a job are finished. However, in many practical environments completed orders are delivered to customers immediately after production stages without any further inventory storage. Therefore, in this paper, we investigate an integrated scheduling model of production and distribution problems simultaneously. It is assumed that products are proceed through a permutation flow shop scheduling manufacturing system and delivered to customers via available vehicles. The objective in our integrated model is to minimize maximum returning time (MRT), which is the time that last vehicle delivers last order to relevant customer and returns to production center. The problem formulated mathematically, and then an improved imperialist competitive algorithm (I-ICA) is proposed for solving it. Furthermore, sufficient numbers of test problems are generated for computational study. Various parameters of the algorithm are analyzed to calibrate the algorithm by means of the Taguchi method. At the end, the effectiveness of the proposed model and suggested algorithm is evaluated through a computational study where obtained results show the appropriate performance of integrated model and solving approach with regard to the other algorithms.

Soft set theory acts as a fundamental tool for handling the uncertainty in the data by adding a parameterized factor during the process as compared to fuzzy as well as intuitionistic fuzzy set theory. In this manuscript, an attempt has been made to compare two intuitionistic fuzzy soft numbers (IFSNs) and then weighted averaging and geometric aggregation operators for aggregating the di erent input arguments have been presented. Further, various properties of its have also been established. The e ectiveness of these operators has been demonstrated through a case study.

In today competitive environment, qualified human resources are considered as one of the major keys to the organizations’ success. So an efficient solution to the problem of personnel selection is more necessary than any time in the past. Besides many of the works in the literature of the field, this paper presents a novel fuzzy ELECTRE approach which is categorized as a multiple criteria decision making (MCDM) technique. In the approach, the weights and ranks are determined by linguistic variables while both quantitative and qualitative criteria are considered simultaneously. At last with a case, the implementation of the model is illustrated and the results are compared with TOPSIS.

Process monitoring is a continuous process for improving the quality. Control chart is a process monitoring tool of SPC tool kit that plays an important role in providing widespread monitoring, to observe the changes in parameters. Mostly the mean control charts are used for monitoring in process location. In a perfect situation, when there are no outliers, the mean charts are more efficient than median control charts. In reality that data is not free from outliers always, so the median charts are considered as the best for monitoring location parameters. The use of an auxiliary variable in a control chart may be the cause of efficiency gain. The current article considers EWMA median charts based on auxiliary variable(s). Different run length performance measures are considered to expedite the proposed charts in both contaminated and uncontaminated process environments under multivariate normal distributions. An illustrative example is provided to validate the performance of proposed charts. From the results, we deduce that the performance of median control charts is much better than mean control charts in the presence of outliers and also the performance of control charts can be enhanced by using more auxiliary variables.

A number of paired comparison (PC) models exists in the literature. In this paper, the posterior distribution of the parameters of the Rayleigh PC model is derived using informative prior: Conjugate and Dirichlet. The values of the hyperparameters are elicited using prior predictive distribution. The preferences for the data of cigarette brands: Goldleaf (GL), Marlboro (ML), Dunhill (DH) and Benson & Hedges (BH) are collected from university students. The posterior estimates of the parameters are obtained under the loss functions: Quadratic Loss Function (QLS), Weighted Loss Function (WLS)and Squared Error Loss Function (SELF) with their risks. The preference and predictive probabilities are calculated. The posterior probabilities, for the hypothesis of comparing two parameters are evaluated. The graphs of marginal posterior distributions are given. Appropriateness of the model is tested by Chi-Square.