فصلنامه چشم انداز مدیریت صنعتی
سال یازدهم شماره 4 (پیاپی 44، زمستان 1400)

Among the decisions related to the milk supply chain, those related to the supply of raw milk from farms to the dairy factories are highly important. In this paper, a two-stage scenario-based possibilistic model is developed for designing a milk supply chain network from farms to the dairy factory in the form of location-routing problem. The milk which is collected by collection center (CC) vehicles or directly is delivered by farmers to CCs. The occurrence of disruption is considered in the form of probable scenarios. A given percentage of capacity of CCs and some of the existing routes might be unavailable under each disruption scenario. A possibilistic programming method is used to cope with epistemic uncertainty in parameters (cost, demand, and milk produced). Because of the mathematical model's high complexity in large sizes, a Lagrangian relaxation algorithm is also devised. The proposed model helps to make optimal decisions in the milk collection process from farms to factories according to existing constraints. The numerical results show the efficiency of the solution approach.

The aim of this study is to present an integrated method to increase the quality of the product during its lifetime through the design of Taguchi experiments and PAF model of quality costs. For this purpose, the combination of the importance of quality costs in the life cycle of the product by combining the PAF quality cost model and designing Taguchi experiments in the Entekhab industrial group in Isfahan has been studied. In this study, four phases of the selected product life cycle (introduction, growth, maturity and decline), at four levels (importance of prevention, evaluation, internal failure and external failure) are considered as control factors. In this study, product quality is considered as a response factor, which is considered to obtain the maximum value. The results of this study on the selected product of Entekhab Industrial Group have been determined. In the introduction stage and growth of product life cycle prevention policy have been selected.  In the maturity and decline stage, the appraisal and domestic failure policy has been chosen as optimal levels. By following these policies, the quality of selected product during its lifetime increases.

In today's societies, given the increasing tendency to buy from malls and megastores that can meet all the buyer's needs in a limited time, we witness a large-scale growth of modern malls. Due to the breadth of services and diversity in such places, it is necessary to have a central intelligent control unit that takes control of all existing systems. The Internet of Things is one of the new technologies in the last decade, playing an essential role in intelligent business computing. The volume of customers coming to the mall, the customer service, and the customer buying behavior are factors whose analyses can dramatically increase the revenue from the stores and lead to better customer satisfaction. Many models are used to serve this purpose, one of which is the Bayesian Network model. In this research, using this model based on the interests and preferences of customers and purchasing patterns, we identify customers' needs and provide them with the desired product. After implementing this model, increasing the potential sales in the store and increasing the performance and speed of the store function are expected.

In this study, a robust two-stage risk-aversion optimization model is proposed for the multi-product relief network design problem. The comprehensive set of decisions for locating and reinforcing relief facilities, inventory planning, and distributing healthcare items has been considered in an integrated manner. Uncertainties of relief facility capacity, relief demand, and the node linkage capacity are considered. Moreover, the weighted average expected loss is considered in the proposed robust planning model. The efficiency of the proposed model has been evaluated by examining numerical instances. The obtained results indicate the efficiency of the distributionally robust model compared to the traditional two-stage stochastic model. In addition, the type of ambiguous set and levels of confidence, risk aversion, and adjustment parameters will affect network performance.

Due to the fact that excellence models, especially EFQM, work based on RADAR logic, which makes them heavily rely on subjective judgments, the aim of this study is to present a more accurate and objective way of assessing organizations with EFQM using Neutrosophic logic approach to minimize estimation errors of assessment values. In this study, assessing organizations based on the EFQM excellence model is performed using Neutrosophic logic, which is a three-valued logic useful in studying uncertainty. First, an assessing tool is designed for evaluation with Neutrosophic logic. Its validity of structure and content was confirmed using EFQM model framework and experts' opinion of evaluation. Also, the reliability was supported in a similar manner using experts' views. Then, the selected organizations, including three state-owned, private, and semi-public banks, were assessed with the assessing tool, and finally, the results of assessing with RADAR logic and Neutrosophic logic were compared. Based on assessing experts of EFQM National Excellence Award and calculations, assessing studied organizations with Neutrosophic logic approach proved more accurate than assessing with RADAR logic approach.

In the pharmaceutical supply chain, pharmaceutical products must be distributed among consumers with good quality at the right time and in the right place. Medicineis a product which affects the health of society and its timely delivery to consumers is of great importance. Therefore, it requires proper planning for its production and distribution. In this paper, we developed a model that minimize the cost of production, inventory, delivery, earliness and tardiness. We also assumed the uncertainty of demand and solved the linear mathematical model using stochastic programming and we solved the problem with stochastic programming. Also, due to the fact that the model with the objective function of earliness and tardiness with different delivery times of NP-hard problem for this problem, a hybrid genetic and variable neighborhood search algorithm were presented. Here, five scenarios were considered, the expected value of perfect information (EVPI) was measured and the obtained results were compared with the two-stage random-scheduling model. The computational results showed the efficiency of the developed model. Also, the results of the proposed hybrid algorithm were compared with the genetic algorithm, and the results showed that in terms of objective function, the hybrid algorithm has a much better performance compared to the genetic algorithm.

In this research, a mathematical model of the vehicle routing problem to support military units is presented and solved. To present this model, first, various criteria extracted from the literature review of vehicle routing issues in the field of military, war and crisis are investigated. Then, the criteria that are important for supporting the military units under study are introduced and the mathematical model of the problem based on these criteria is presented. One of the salient features of the current research compared to similar researches is the simultaneous consideration of five effective criteria in supporting the units of this organization, which include "time window for delivery of goods to units", "ability to pick-up and deliver goods on the road transport route", "the heterogeneity of the fleet of road vehicles", "the need to send goods from multi-depot" and "the need to transport several types of goods". Since this is one of the optimization problems in the family of NP-hard problems, GA, PSO and SA algorithms were used to solve the model. In order to validate, the results of these algorithms have been compared with the exact solution results with GAMS software, which shows the proper performance of the proposed genetic algorithm.

Nowadays, recommender systems have reshaped the ways of information filtering between websites and the users in order to identify the users’ interests and generate product suggestions for the active users. Recommender systems are generally divided into three groups: Content-based, Knowledge-based, and collaborative-based, and in some cases hybrid. The main idea of collaborative filtering is that they predict a user’s interest in new items based on the recommendations of other people with similar interests. This Approach does not require having knowledge about items. Collaborative filtering has two main types: Memory-based and Model-based. Memory based Collaborative filtering makes use of user rating dataset to compute similarity index between set of users or set of items. The main purpose of this article is to offer a Memory-based Collaborative recommender system in order to optimize the results of Collaborative filtering algorithm. In the proposed method, the combination of fuzzy Grey Wolf Optimizer algorithm and Lion Optimization Algorithm is used to find the most similar users to the target user. The results of the proposed method confirmed a significant increment in Precision, Recall and F-measure in comparison with baseline methods.

The classical economic production quantity (EPQ) model was developed to manage inventory costs in companies last decade ago. This model is extended in various directions in recent years. The classic EPQ has some unrealistic assumptions. The model assumes that all products are perfect, while the production of defective items is inevitable in the real-world environment. Another assumption relates to the continuous demand satisfaction, which ignores the commonly used multiple shipments policy in practice. Finally, the classic model does not consider the probabilistic breakdown of the machine and the required maintenance activities. The present work aims to develop a new imperfect EPQ model under probabilistic machine failure, corrective maintenance, and multiple shipments policy. Two cases are investigated: 1- Considering the production of a fixed percentage of imperfect items 2- Considering no production of defective items. Due to the complexity of the problem, a numerical bisection method is utilized to solve the problem and finding the best possible production time. This method's performance is evaluated by comparing it to the obtained solutions by MATLAB optimization toolbox for genetic and simulated annealing algorithms. Sensitivity analysis is performed, and finally, some directions for future research are suggested.

The purpose of this research is to evaluate the level to which a company’s activities in a supply chain are LARG. In this study, an integrated method is used to evaluate the LARG supply chain performance of a company resulting from the integration of LARG concepts and Balanced Scorecard approach. The BSC measures are selected based on the LARG concepts, and then the indicators entered into the dynamic model. Variables are changed in different scenarios to analyze changes in the company’s performance. Scenarios are designed to evaluate the supply chain performance using the strategic objectives. The results show that simultaneous implementation of LARG elements is not possible due to the trade off relationship. By analyzing the scenarios, it was found that by changing each parameter in the dynamic model, some LARG elements increase and at the same time, some other elements decrease. For example, by increasing the productivity of education, the level of leanness and resilience increases, but it has no effect on the environment. Using the designed dynamic model, the effect of each managerial action and decision on LARG can be determined and the extent to which strategic goals can be achieved.