Abstract
Performance-related indicators are crucial for evaluating and forecasting performance, enhancing decision-making efficiency, and establishing sustainable growth strategies. They motivate individuals and organizations, increase transparency, and accurately measure organizational performance, enhancing cohesion and resource development. In this paper, the performance of the organization is investigated. In the first phase, the data is preprocessed and normalized, and then, in a three-stage process, the index selection is performed. In the first stage, Subtractive Clustering is used for categorizing the samples. Then, using the Silhouette index, the quality of clustering is evaluated. In the second stage, each candidate index is assigned a rank, and a one-dimensional convolutional neural network is used to predict the organization's performance based on the selected indices. The parameters of the convolution and pooling layers of the neural network are adjusted using a learning automata model, and finally, the selection operation is performed using the specified ranking by the FSFS method. The experimental results show that the proposed method achieved an average accuracy of 88.12% during the company revenue evaluation stage and a higher accuracy of 93.12% in the assessment of customer satisfaction, demonstrating superior performance.