In the complex and dynamic stock market landscape, investors seek to optimize returns while minimizing risks associated with price volatility. Various innovative approaches have been proposed to achieve high profits by considering historical trends and social factors. Despite advancements, accurately predicting market dynamics remains a persistent challenge. This study introduces a novel deep reinforcement learning (DRL) architecture to forecast stock market returns effectively. Unlike traditional approaches requiring manual feature engineering, the proposed model leverages convolutional neural networks (CNNs) to directly process daily stock prices and financial indicators. The model addresses overfitting and data scarcity issues during training by replacing conventional Q-tables with convolutional layers. The optimization process minimizes the sum of squared errors, enhancing prediction accuracy. Experimental evaluations demonstrate the model's robustness, achieving a 67% improvement in directional accuracy over the buy-and-hold strategy across short-term and long-term horizons. These findings underscore the model’s adaptability and effectiveness in navigating complex market environments, offering a significant advancement in financial forecasting.

Over the past three decades, the unified approach of the optimization of the logistic systems has become one of the most important aspects of optimizing the supply chain so that in recent decades it has had a large application in practice and has been used to increase the efficiency and effectiveness of the logistic fleet. The inter-urban transport networks by the terminals play an important role in logistic fleet and the goods distribution. In some cases, numbers of terminals face with an overload and others encounter with additional vehicles, which result delay in the load post and unnecessary car downtime. The present paper aims at modeling, scheduling and routing of the vehicles network and minimizing delays in order to create an optimal balance between the number of vehicles and the capacity of the car terminals to use the maximal capacity of vehicles. So that the multi-objective mathematical model is presented to quantify the regular transportation costs and to minimize the car downtime. The proposed model has two conflicting objectives where on tried to increase costs and the other decreases unused cars. Due to the high complexity of the problem, the multi-objective differential evolutionary algorithm (MODE) has been used. To prove the proposed algorithm, it has compared with the NSGA-II algorithm using four comparing indexes. The computational results show the superiority of the proposed algorithm.

One of the important factors for the success of a bank industry is to monitor their customers & apos;
behavior and their point-of-sale (POS). The bank needs to know its merchants & apos; behavior to find
interesting ones to attract more transactions which results in the growth of its income and
assets. The recency, frequency and monetary (RFM) analysis is a famous approach for
extracting behavior of customers and is a basis for marketing and customer relationship
management (CRM), but it is not aligned enough for banking context. Introducing RF*M* in
this article results in a better understanding of groups of merchants. Another artifact of RF*M*
is RF*M* scoring which is applied in two ways, preprocessing the POSs and assigning
behavioral meaningful labels to the merchants’ segments. The class labels and the RF*M*
parameters are entered into a rule-based classification algorithm to achieve descriptive rules of
the clusters. These descriptive rules outlined the boundaries of RF*M* parameters for each
cluster. Since the rules are generated by a classification algorithm, they can also be applied for
predicting the behavioral label and scoring of the upcoming POSs. These rules are called
behavioral rules.