The grading of bank branches is a costly and time-consuming auditing procedure that requires the evaluation of multiple qualitative and quantitative criteria. This study proposes a data-driven decision-support framework for extracting decision rules under uncertainty and applying them to infer the future grade of each branch based on selected performance indicators. In the first stage, the branch grading results obtained over a five-year period are treated as the decision attribute, while the corresponding financial indicators are used as condition attributes within the Rough Set Theory framework. This stage identifies the most accurate rule-based classification model and determines its reduct as the most relevant subset of indicators for branch grading. In the second stage, the future values of the selected indicators are forecasted for the subsequent two years. Given the limited size of the available data, several grey forecasting models are employed and compared. Finally, the decision rules extracted in the first stage are applied to the forecasted indicator values to infer the future grades of bank branches with high accuracy, thereby reducing dependence on extensive and time-consuming auditing procedures. The findings demonstrate that combining Rough Set Theory with Grey forecasting models offers an effective predictive and knowledge-based decision-support tool for evaluating branch performance in dynamic and uncertain environments. The proposed framework can support banking managers and policymakers in performance assessment, branch network reorganization, strategic planning, and evidence-based decision-making.
Last-mile delivery planning depends on short-term demand forecasts and vehicle routing, but the step that turns a forecast into a fleet and a routing instance is rarely specified or evaluated. We present FALCON, a two-stage platform in which a rule-based supervisor classifies delivery zones as forecastable, baseline or rejected; a transfer-learned long short-term memory (LSTM) network with multi-head attention produces seven-day hourly forecasts with an upper bound calibrated on a held-out period; a sequential heterogeneous fleet-sizing rule converts payload and stop-time limits into orders per delivery wave; and a fully specified vehicle routing problem with time windows is solved by a genetic algorithm, particle swarm optimisation and continuous-domain ant colony optimisation sharing one random-key encoding, decoder, fitness and a repair with a proved coverage guarantee. On the public LaDe corpus (1.86 million orders, four other cities for pre-training, a leakage-free chronological protocol), the supervisor avoids large errors on sparse zones (99
This paper exhibits a novel method for addressing the multi-objective solid profit transportation problem. Our two objectives, maximisation of profit and minimization of time, are inherently incompatible. So, instead of addressing simply the membership function, as most research do in fuzzy sense, we employ intuitionistic fuzzy set theory to resolve the conflict nature of the objectives by considering both membership and non-membership functions. Because it maximises the degree of satisfaction while minimising the degree of dissatisfaction of an intuitionistic fuzzy decision, the results are preferable to those obtained by only considering the membership functions. Furthermore, a membership function and a non-membership function are developed in terms of both optimistic and pessimistic points of view for each objective. All of the parameters are designed to be parabolic trapezoidal fuzzy in nature due to instinctive imprecision. To obtain the crisp model, a defuzzification approach is utilised. This proposed approach utilizes a combination of the intuitionistic fuzzy sets concept with the fuzzy interactive satisfied method. It assists decision makers in problem solving by providing the reference value of both the membership and non-membership functions in order to reach the ideal value. The computational findings clearly show that, as compared to the solutions acquired in a pessimistic view, the solutions obtained in an optimistic view are more favourable, which is obvious. In this proposed method, the decision makers can vary their target level by adjusting the reference values and on the basis of their satisfaction levels they can pick their desired results. However, the technique proposed in the given paper is much more practical and simpler to use.
This paper presents an inexact general multi-objective linear programming problem and its solution procedure. The coefficients and variables involved in the problem are inexact numbers. The method combines the existing general multi-objective linear programming and interval programming techniques. The model is thus constructed by utilizing linear programming and enabling uncertainty to appear in the coefficients of the objectives and constraints as intervals under the optimization structure. The proposed method of solution elicits an optimal solution with the highest number of objectives attaining their ideal values. The solution is also presented in inexact form. The solution procedure of the proposed technique has been illustrated with the help of a multi-objective portfolio selection problem with both maximizing and minimizing types of objectives.
The selection of suppliers is one of the most vital strategies for improvement of the quality of the output in many organizations. Throughout this process, a buyer aims to select the proper supplier(s) among all the possible choices, using evaluation criteria and various techniques. One of the methods to evaluate and select suppliers is the application of data envelopment analysis (DEA), whose importance in supply chain management has attracted much attention in the past decade. In this research, we develop a model that combines DEA and game theory and enables decision-makers to select suitable suppliers among the available alternatives. The inputs and outputs of the model include social and environmental criteria along with economic criteria. A conservative approach is used to support decision-making under various conditions. A game-theoretical approach is also used to optimize the order size for each supplier in order to maximize the profit of the whole supply chain. We also present the result of a case study that was initiated and performed in one of the stone manufacturing and supply companies in Iran. The data collected for this case study is related to 19 stone mines as suppliers, 10 of which are selected and offered to the manufacturing company. Using a game-theoretical approach, the optimal size of the order to be placed at each supplier is also provided for the stone manufacturing unit.
