Ali AlArjani

Industrial IoT maintenance often stops at health prediction, leaving maintenance, rescheduling, safety, and communication to separate decision processes. This study presents MIRA, a safety-constrained graph-based multi-agent reinforcement learning architecture for joint prescriptive maintenance, production rescheduling, and event-triggered communication. Machine condition was estimated from CNC milling data using temporal convolutional models; because predictive uncertainty failed a predefined validation gate, the controller used deterministic health estimates. Evaluation covered five controllers, six simulated scenarios, and 1800 matched episodes. Relative to Graph-MAPPO, MIRA reduced operational cost by 9.38%, weighted tardiness by 28.10%, unexpected failures by 17.39%, message count by 84.98%, and transmitted data by 83.83%, while increasing on-time completion by 23.55%, without a detectable difference in corrected critical-message recall. Across the three independently trained seeds, failures, safety violations, and message count favored MIRA consistently, whereas cost and tardiness favored MIRA in two seeds. Disabling the execution shield increased safety violations from 0 to 3.56 per episode. Post-training variation in the projected-health safe-start threshold from 0.124 to 0.132 produced no safety violations and only small changes in aggregate operational outcomes. Cross-domain health transfer to PHM 2010 failed without adaptation. The results support simulator-level decision coordination, while broader replication, variable-size deployment, and factory validation remain necessary.

The process of addressing customer requests is a strategy designed to fulfil them in time; it must be free of lateness or negligible shortfalls, as these will lead to higher expenses for the business. Consequently, it must carry out a whole aggregate request handling to tackle a determined number of customer requests (Kanban) according to the capacity owned by the communication companies (i.e., scheduling process). Digital twin (DT) has been tailored to simulate the efficiency of hybridization of ARIMA and SR-tree technique in forecasting, handling, and fulfilling the customers' requests (edges) via Three-echelons Flying Logistic System (TEFLS), where echelons consist of multi-suppliers (e.g., Base-Station Antenna Towers), a manufacturer (e.g., UAVs that handle the customer requests), and multi-customers (i.e., who create the requests). The problem lies in covering service in zones that have weakened signals of base-station towers due to sprawling mountain areas. Therefore, using UAVs to be mediators to load the signal from the nearest tower to connect with any requested edge. The DT helps in determining the diameter of UAVs service zones, the capacity of UAVs handling requests in time, and the handling policy that follows Multitasking Echelons (MTE) toward Multitasking optimization (MTO) to be high performing. UAVs receive four types of customer’ requests, each of which has subclasses. First are regular telephony services that embed emergency calls that improved by approximately 63%, while the second service is a download/upload service and includes GPS location search that improved by 72% and therapeutic alerts according to the aforementioned request from the customer to receive the service as contracted with the organised company, which improved by 93%. Finally, there is browsing service and online business that improved by 89%.

Forecasting models are essential instruments for a nation's economic growth and policymaking. This study's primary objective is to predict Saudi Arabia's CO2 emissions. Holt-Winters Exponential Smoothing (H-W), Autoregressive Integrated Moving Average (ARIMA), and Artificial Neural Network (ANN) models are used in this study to forecast Saudi Arabia's CO2 emissions. This study forecasts time series data on the Kingdom of Saudi Arabia's CO2 emissions from 1960 to 2014 using ANN, H-W, and ARIMA models utilizing statistical software. Furthermore, this study uses a variety of accuracy metrics to demonstrate the forecast model's performance. The Kingdom of Saudi Arabia's CO2 emissions can be predicted using the ARIMA (2,1,2) model. This study will assist the researcher in better understanding CO2 emission forecasts. In addition, government entities can use the findings of this study to establish strategic plans.

The efficiency of the supply chain is highly corelated with the acceleration of preparing orders from the warehouse to meet customer requests on time. This objective can be activated by guiding via UAV to sketch the shortest route collect all customer requests’. A unified framework has been built consisting of three sequential stages guaranteeing quick customer responses to meet SDG 8.0 about economic growth via Working Capital Turnover. The preamble stage aims to sort the items according to XYZ/ABC pattern. The 2^nd stage classifies the customers’ orders according to similarity via hybridising the (Extreme Gradient Boosting) XGboost and GRU (gate recurrent unit) algorithms that generate all possible orders’ lists in 1.2586 sec with RMSE 0.01395, which is filtered in the 3^rd stage via creating their shortest route via hybridising (Ant Colony Optimisation) ACO and GRU algorithms to give workers to pick up more items in minimum time to accelerate preparing orders. The challenge lies in assigning a new order to one of the lists that belongs to Map (A) and creating Map (B_i). The paper presents innovative tuning parameters for ACO that help in accelerating orders preparation by 152.49%/worker/day and achieve overall equipment effectiveness (OEE) approximate to 88%. The ACO_GRU superior to Regular Ant Colony (RgACO), Elitist Ant Colony (EACO), Ranked Ant Colony (RnACO), Best-Worst Ant Colony (BWACO), Min-Max Ant Colony (MMACO), Mat-ACO, A* algorithm, Dijkstra’s algorithm, and MRL-SA with 19%, 16%, 9%, 12%, 6%, 6%, 5%, 5%, 3%, respectively. Therefore, the proposed model called “Acceleration Customer Orders’ Preparation” (ACOP).

This paper formulates six inventory models for products with instantaneous deterioration, focusing on the impacts of full and partial advance payment structures. The demand function depends on both price and stock levels and accounts for shortages through full backlogging. The primary objective is to determine the optimal payment policy under varying trade facilities, analyzing six distinct payment scenarios commonly employed in business practice. Each model is presented with closed-form solutions and supported by mathematical formulations. For each case, algorithms and mathematical proofs are developed to determine the optimal cycle duration and corresponding unit cost. Numerical examples and 2D graphical representations generated using MATLAB are included to validate the proposed models. Additionally, a sensitivity analysis is conducted to examine the effects of each payment policy and parameter variation, providing key managerial insights into payment planning in inventory management.

Selecting a supplier is a critical strategic decision for supply chain management in today's global context. The process involves evaluating suppliers based on core competencies, pricing, delivery timeframes, location, data gathering, and related risks. Suppliers play a crucial role in an organization's profitability and stability. Finding the most optimal supplier can help industries reduce material expenses and maintain their competitive advantage. The supplier not only impacts the organization's profit margin but also its economic strength. Choosing a supplier requires considering qualitative and quantitative elements, making it a decision issue with several criteria. This study aims to create and evaluate a supplier selection model using the analytical hierarchy approach, focusing on a specific case study. When selecting the best supplier, it is crucial to consider tangible and intangible elements that may conflict. The supplier selection process considers several criteria, including qualitative and quantitative variables. The proposed methodology involved a literature review and informal interviews with industry experts and academics to establish the selection criteria. "Quality Supplier Co." was chosen due to the paramount importance of their quality. This research will comprehensively analyze several criteria to identify suppliers accurately.

In the struggle to achieve environmental management goals, wind energy is essential. On the other side, the effectiveness of the wind turbines is a prerequisite for wind energy to reach its full potential. The purpose of the research is to examine how artificial intelligence (AI) techniques are applied to wind speed. In the context of wind speed forecasting, three network parameter optimization algorithms—AdaGrad, RMSprop, and Adam—are applied and contrasted in this work. The wind speed data used in this article came from the Saudi Arabian Jeddah Al Jazeera database. The four metrics used to evaluate performance are mean absolute error (MAE), mean square error (MSE), root mean square error (RMSE), and R-squared. According to the experiment's findings, the Adam algorithm performs better than the other optimization techniques.

Projects, as catalysts for proactive transformation, offer a temporary and adaptable framework that effectively handles complexities (or uncertainties) within a competitive corporate landscape. Hence, the use of an effective project management framework, such as Dynamic Project Scheduling (DPS), is a method to handle intricacies in order to accomplish organizational objectives. DPS refers to a triangle interaction involving baseline scheduling, schedule risk analysis, and project control while supporting schedule adjustment in response to changes and uncertainties. However, there is a lack of information regarding studies that have investigated the feedback mechanisms among DPS components. This study was designed to examine the counterintuitive relationships between these components using system dynamics. The quantities within the DPS system were identified and defined. A casual loop diagram was used to illustrate the interactions among these quantities. Subsequently, a Stock and Flow Diagram (SFD) was created to identify the inputs, states, and flow mechanisms within the DPS. Using the SFD, a system dynamics expression was generated which was then employed to compute the rate of change of the Budgeted Cost of Work Remaining (BCWR) for two projects at different time intervals. The results properly indicated the period of idleness during project execution. The use of BCWR rather than schedule variance provides a more effective visual representation for evaluating performance and tracking progress. The BCWR and planned value exhibit contrasting trends, highlighting the importance of earned value quantities in project control. The use of system dynamics in project management can enhance the planning and scheduling phase, allow project managers to monitor pertinent performance measures, and optimize project outcomes through informed decisions.

This paper focuses on the production systems that may produce a proportion of recyclable defective products. The developed model is called an Economic Recycle Quantity (ERQ) model with the assumption of a full recovery of defective items. The defective parts are collected during the production-off time and can be used during the next production cycle of the same category. The demand rate of the non-defective items is a two-level piecewise factor—one during the production-run time and another during the production-off time. The developed model aims to optimize the total inventory cost, the order quantity, and the amount of recyclable defective items that represent the ERQ. The mathematical formulations of the model are deduced theoretically. The model was solved analytically, and the optimal results are illustrated. Sensitivity analysis is carried out to investigate the effect of varying system parameters and validate the proposed model. Results of the sensitivity analysis show that the consideration of defective part recycling reduces the total inventory cost where the raw material is reduced. The cost reduction is about 1%; of course, the environmental impact is more appreciated. Furthermore, the managerial implications are described, and the future perspectives are discussed.