Scientia Iranica
Volume:28 Issue: 6, Nov-Dec 2021

This study proposes a framework for supplier evaluation, selection, and assignment that incorporates a two-stage game-theoretic approach method. The objective is to provide insights to manufacturers in choosing suitable suppliers for different manufacturing processes. The framework applies to the decision logic of multiple manufacturing processes. In the first stage, a non-cooperative game model is utilized for supplier evaluation and selection. The interactive behaviors between a manufacturer and some supplier candidates are modeled and analyzed so that the supplier evaluation value (SEV) can be obtained using the Nash equilibrium. In the second stage, the supplier evaluation values become the input for the Shepley values calculation of each supplier under a cooperative game model. The Shapley values are utilized to create a set of limited supplier allocation. This paper provides managerial insights to verify the proposed approach on supplier selection and allocation. Thus enables SCM manager to optimize supplier evaluation, selection, and order assignment.

Skewness plays a vital role in different engineering phenomena so it is desired to measure this characteristic accurately. Several measures to quantify the extent of skewness in distributions have been developed over the course of history but each measure has some serious limitations. Therefore, in this article, we propose a new skewness measuring functional, based on distribution function evaluated at mean with minimal assumptions and limitations. Four well recognized properties for an appropriate measure of skewness are verified and demonstrated for the new measure. Comparisons with the conventional moment-based measure are carried out by employing both functionals over range of distributions available in literature. Furthermore, the robustness of the proposed measure against unusual data points is explored through the application of influence function. The Mathematical findings are verified through meticulous simulation studies and further verified by real data sets coming from diverse fields of inquiries. It is witnessed that the suggested measure passes all the checks with distinction while comparing to the classical moment-based measure. Based on computational simplicity, applicability in more general environment and preservation of c-ordering of distribution, it may be considered as an attractive addition to the family of skewness measures.

In this paper, a novel method is proposed to predict the cost of short-term hourly electrical energy based on combined neural networks. In this method, the influential parameters that play a key role in the accuracy of these systems are identified and the most prominent ones are selected. Due to the fluctuations of electricity prices during various seasons and days, these parameters do not adhere to the same pattern. In the proposed method, initially, using the SOM network, similar days are placed in close clusters. In the next stage, the temperature parameter and prices pertaining to similar days are trained separately in two MLP neural networks because of their differences concerning the range of changes and their nature. Finally, the two networks are merged with another MLP network. In the proposed hybrid method, an evolutionary search method is used to provide an appropriate initial weight for neural network training. Given the price data changes, the price amidst the previous hour has a significant effect on the prediction of the current state. In this vein, in the proposed method, the predicted data in the previous hour is considered as one of the inputs of the next stage.

This paper presents a study over the behavior of six air pollutants including PM10, PM2.5, O3, SO2, NO2 and CO in Tehran during a 6-year timespan. In this paper, an iterative procedure based on the univariate Box-Jenkins stochastic models is applied to develop the most effective forecasting model for each air pollutant. Applying a number of widely used criteria, the best model for each air pollutant is selected and the results show that, the proposed models perform accurately and satisfactorily for both fitting and predicting where, the fitted and predicted values are so close to the true values of the related data. Finally, a factor analysis is conducted to investigate the relationships between the air pollutants where the results show that four factors accounts for 93.2704% of the total variance. In this regard, the factor containing PM10 and PM2.5 and the factor containing CO and NO2 are, respectively, the most and the second most affecting factors with proportion of 43.2594% and 21.6500% of total variability. While both factors originate from high number of automobiles which use fossil fuels, decreasing the number of automobiles or increasing the quality of fossil fuels may result in up to 60% improvement in air quality.

Nowadays, global competition causes that the companies are dealing with the issue of cost reduction besides increasing productivity in business network more than ever. Because of that, today, both researchers and industrial practitioners are focusing on the supply chain network issues. In order to achieve the real word objectives, we attempt to improve the efficiency of a supply chain via not only considering simultaneous pick up and split delivery but also minimizing the total costs and maximizing the customer service in the form of multi-products and multi-period. In addition, to accumulate the data of parameters, a case study in a food industry in the north of Iran has been utilized. Eventually, the proposed mixed-integer linear programming model is addressed by a ε-constraint approach. Finally, related results of this solution are analyzed and also is compared with simple VRP.

The ratio, product, and regression estimators are commonly constructed based on the conventional measures such as mean, median, quartiles, semi-interquartile range, semi-interquartile average, coefficient of skewness, and coefficient of kurtosis. In case of the presence of outliers, these conventional measures lose their efficiency/performance ability and hence are of less efficient as compared to those measures which performed efficiently in the presence of outliers. This study offers improved class of estimators for estimating the population variance using robust dispersion measures such as probability-weighted moments, Gini’s, Downton’s and Bickel and Lehmann measures of an auxiliary variable. Bias, Mean square error (MSE) and minimum MSE of the suggested class of estimators have been derived. Application with two natural data sets is also provided to explain the proposal for practical considerations. In addition, a robustness study is also carried out to evaluate the performance of the proposed estimators in the presence of outliers by using an environment protection data. The results reveal that the proposed estimators perform better than its competitors and are robust, not only in simple conditions but also in the presence of outliers.

This paper presents a new redundancy allocation problem for a system with the k-out-of-n configuration at the subsystems’ level with two active and cold standby redundancy strategies. The failure rate of components in each subsystem depends on the number of working components. The components are non-reparable, and the failure rate of the component can be decreased with some preventive maintenance actions. The model has two objective functions: maximizing the system’s reliability and minimizing the system’s costs. The system aims to find the type and number of components in each subsystem, redundancy strategy of subsystems, as well as the decreased values of components failure rates in subsystems. Since the redundancy allocation problem belongs to NP-Hard problems, two Non-Dominated Sorting Genetic Algorithm II (NSGA-II) and Non-Dominated Ranked genetic algorithm (NRGA) metaheuristic algorithms were used to solve the presented model and to tune algorithms parameters we used response surface methodology (RSM). Besides, these algorithms were compared using five different performance metrics. Finally, the hypothesis test was used to analyze the results of the algorithms.

This paper analyzes the optimal price and quality decisions of a retailer for its differentstores in a heterogeneous market. The consumers are assumed to be heterogeneous in theirwillingness to pay for quality and are non-uniformly distributed in the market. This type ofheterogeneity which is identified based on income disparity can have important implicationsfor a retailer’s optimal policy. The specific objective of this paper is to investigate how thedistribution of consumers’ types in the market and their travel costs affect the optimal settingof price and quality levels among different stores of a retailer. Our results express that thegeographical disparity of willingness to pay plays a significant role in the differentiation andtargeting strategy of a retailer. Comparative analysis shows that the widely adopted assumptionof uniform distribution of consumers in the literature leads to non-optimal decisions where thedistribution of consumers is non-uniform in a real-world situation.

During the years of imposed sanctions against Iran, Iran Khodro Company (IKCO) got into a hazardous situation due to CKD parts’ purchasing cost increment and emersion of new product variants in the competitive market. To examine such situation, this study examines a multi-period semi-centralized dual-channel supply chain where a common retailer (free market) and two manufacturers’ (IKCO and Saipa as a major competitor) direct channels are confronted with reference price dependent and stochastic demand. The problem is analyzed under Stackelberg and cooperative games scenarios using heuristic algorithm and a League Championship algorithm respectively, as solution methods. Results obtained from solving the problem with IKCO data proves higher profitability of the cooperative game and its remarkable resilience for all products’ memory types i.e. short/long term memory against production cost disruption which is imposed to IKCO in some periods. Besides calculating Saipa’s optimal wholesale price in the disruption periods, our approach with support of experimental analyses is able to assign a supply chain’s degree of resilience against disruptions to its product’s memory type and also power structure.

Computer numerical control (CNC) is a manufacturing concept where machine tools are automated to perform some predefined functions based on the instructions fed to them. CNC turning processes have found wide ranging applications in modern day manufacturing industries due to their capabilities to produce low cost high quality parts/components with very close dimensional tolerances. In order to exploit the fullest potential of a CNC turning process, it should always be operated while setting its different input parameters at their optimal levels. In this paper, two classification tree algorithms, i.e. classification and regression tree (CART) and Chi-squared automatic interaction detection (CHAID) are applied to study the effects of various turning parameters on the responses and identify the best machining conditions for a CNC process. It is perceived that those settings almost match with the observations of the earlier researchers. The CART algorithm outperforms CHAID with respect to higher overall classification accuracy and lower prediction risk.

The main activity in postal services is to deliver letter mails and parcels. By changing customers’ needs and behaviors along with emerging new technologies, postal services have to be renovated. Thus, understanding the changing environment, forecasting the performance, identifying key drivers, and making effective interventions are critical for any further actions. Performing these actions for Iran Post Company is the focus of this paper. Therefore, system dynamic approach is chosen, effective variables are determined and causal-loop diagram (CLD) and stock and flow diagram (SFD) are developed. The results are then validated using expert panels and historical data and the developed model is utilized for policy making. Therefore, two scenarios are designed based on changes in postal rates, quality of services and e-service market share. These scenarios could provide CEOs with critical information to make effective interventions.

Efficient management of providing home health care services requires many considerations. In this paper, a mathematical model for the daily staff routing and service scheduling is developed for home health care companies. In this model, both economic factors and qualitative service-oriented performance measures are simultaneously optimized. To make the model more realistic, many real situations such as considering different qualifications and diverse vehicles for staff members, different requirements and predetermined preferences for patients, possible temporal interdependencies between services, and continuity of care (CoC) are taken into account. We also added some important constraints related to blood sampling requirements, which make our proposed model more complex. The proposed model is a mixed integer linear programming model (MILP) that belongs to an NP-hard class of optimization problems. To solve such a complex mathematical model, a genetic algorithm (GA) is proposed to find near-optimal solutions. We use some randomly generated test instances with different sizes to evaluate the performance of the GA. Finally, it is demonstrated how the proposed solution scheme can end up with proper scheduling and routing policies compared to those obtained through exact methods.

The purpose of this paper is to introduce a generalization of Molodtsov's approach to soft sets obtained by passing from the classical two-valued logic underlying those sets to a three-valued logic, where the third truth value can usually be interpreted as either non-determined or unknown. This extension of soft set approach allows for more intuitive and clearer representation of various decision making problems involving incomplete or uncertain information. In other words, it is a useful way to analyze soft set based multi-criteria group decision making problems under the lack of information resulting from the inability to determine the data. In this paper, we introduce the concept of three-valued soft set and its some basic operations and products. We propose the formulas to calculate all possible choice values for each object in the (weighted) three-valued soft sets, and thus calculate their respective decision values. By modifying the TOPSIS and ELECTRE methods to deal with multi-criteria group decision problems, three-valued soft set based decision making algorithms are constructed. To demonstrate the practicality of these algorithms, we address the outstanding examples adapted from the decision problems in real-life. Lastly, some aspects of the efficiency of the proposed algorithms are discussed with computational experiments.