Equipment Scheduling Research Articles

Occupancy and equipment usage prototype schedules for building energy simulations of office building types in China

The accuracy of occupancy and energy-consuming equipment schedules significantly influences building energy simulations. Existing standards provide generalized schedules that do not fully capture variations across different office building types and periods. This study utilizes questionnaire data from office buildings across four Chinese cities to extract refined prototypes for occupancy and equipment usage schedules using hierarchical clustering analysis. Specific schedules are developed for summer workdays, winter workdays, summer weekends, and winter weekends, covering HVAC, lighting, and office equipment. For instance, the extracted lighting schedule increases annual energy consumption intensity by 25.35% compared to standard schedules. Additionally, XGBoost models identify key factors influencing equipment usage; for HVAC, floor number emerges as most significant. The study's prototypes offer more realistic inputs for building energy simulations, enhancing accuracy and guiding energy-efficient building design and management strategies in China.

Research on multi-time scale Volt/VAR optimization in active distribution networks based on NSDBO and MPC approach

To explore the potential of all-round and multiangle voltage and reactive power control to reduce losses in a new power system with a high proportion of distributed resources connected to a distribution network, this study proposes a Volt/VAR optimization method for distribution networks using a nondominated sorting dung beetle optimizer (NSDBO) and model predictive control (MPC). According to the characteristics of the various adjustable resources in the network, they are divided into day-ahead and intraday optimizations. The optimization process in the day-ahead stage uses the NSDBO to create a discrete equipment scheduling plan. The optimization process in the intraday stage combines the discrete equipment scheduling plan with the MPC to create the scheduling plan for photovoltaics, energy storage, and vehicle charging stations. Two-stage optimization scheduling is achieved with the lowest discrete equipment regulation cost and minimization of network loss and node voltage deviation penalty cost as the optimization goal. The feasibility of this method for coordinating various resources in the network, processing discrete and continuous variables, and coping with the volatility and uncertainty of high-proportion distributed resources is verified through case analysis. The effectiveness of this method in improving the security and economy of distribution networks is demonstrated. The superiority of the solution speed and quality is also confirmed.

The Influence of Intelligent Guided Vehicle Configuration on Equipment Scheduling in the Railway Yards of Automated Container Terminals

The Influence of Intelligent Guided Vehicle Configuration on Equipment Scheduling in the Railway Yards of Automated Container Terminals

AI learning-driven optimization of microgrid systems for rural electrification and economic empowerment

Artificial Intelligence (AI) and machine learning (ML) are transforming the landscape of rural electrification through their application in microgrid systems. Microgrids, localized networks that can operate independently or in conjunction with the main grid, offer a viable solution for delivering reliable electricity to rural areas. AI-driven optimization enhances the performance and efficiency of these microgrids, addressing the unique challenges faced in remote regions and contributing to economic empowerment. AI technologies, including machine learning algorithms and data analytics, play a critical role in optimizing microgrid systems. By employing predictive maintenance, AI can forecast equipment failures and schedule timely repairs, thereby minimizing downtime and reducing maintenance costs. Additionally, AI enhances energy management through precise load forecasting, optimizing energy distribution, and integrating renewable energy sources like solar and wind. This not only balances energy supply and demand but also mitigates the variability associated with renewable sources. The impact of AI-driven microgrid optimization on rural electrification is profound. Improved system reliability and efficiency result in more stable energy supplies, which is crucial for supporting local businesses, educational facilities, and healthcare services in remote areas. Economically, the reduction in energy costs and the creation of new job opportunities further empower rural communities. Socially, reliable electricity enhances the quality of life, enabling better educational outcomes and promoting economic development. Successful implementations of AI in microgrid systems have demonstrated significant benefits. For instance, AI-optimized microgrids in rural communities have shown improved performance and stability, leading to greater energy access and economic benefits. However, challenges remain, including technical complexities, financial constraints, and regulatory hurdles. Addressing these challenges through continued research, investment, and policy support is essential for scaling AI-driven solutions and achieving broader impacts. AI-driven optimization of microgrid systems offers a powerful tool for advancing rural electrification and economic empowerment. By leveraging AI technologies, we can enhance energy reliability, reduce costs, and foster economic growth in underserved regions, paving the way for a more equitable and sustainable future. Keywords: AI Learning-Driven, Microgrid Systems, Rural Electrification, Economic Empowerment.

DIGITIZATION AND INNOVATION IN HEALTHCARE SERVICES

Innovation and digital technologies have caused major changes in most fields today, including health services. Thus, access to telemedicine, investigations with modern equipment and online scheduling improved the time spent by patients in receiving the services they needed. In this context, in the last two decades, the interest of academics and specialists in studying the complex relationship between digitization and innovation in the medical field has increased. Therefore, the purpose of this research is to highlight the conceptual links between "healthcare", "innovation" and "digitalization" in order to highlight the current concerns of this sector. Using a bibliometric analysis and the Web of Science database, the paper manages to analyze 650 scientific documents in which these concepts were found. The results of the survey highlight the fact that there was a high interest in research on the relationship between these concepts during the pandemic period. Also, the thematic map created identified specific clusters related to digitization and health. From a theoretical point of view, the research complements academic studies in this field and in particular extends bibliometric theories. From a practical point of view, this study provides valuable information to practitioners and decision-makers in the medical field, and especially to those who are in full development and digitalization process. As an element of originality, the paper identifies thematic groups that provide readers with information on emerging topics in this field.

A Profit Maximization Problem of Equipment Rental and Scheduling with Split Services

This study addresses a scheduling problem arising from the “Rent Instead of Buy” business model where rental service platforms rent equipment from suppliers to satisfy customer requests in exchange for a lump-sum payment. Suppliers often implement piecewise linear concave charging policies to encourage longer rental duration. This policy prompts the platform to consolidate service periods into longer rental agreements for lower rental rates, potentially necessitating the splitting of a customer request among multiple facilities. We establish an integer programming model to maximize the platform’s profit, which is equal to the total revenue from serving customers minus the total rental and scheduling costs. To solve the problem, we propose two Dantzig-Wolfe decomposition reformulations and design branch-and-price algorithms accordingly. Furthermore, by exploiting the property of the concave rental cost functions, we derive the optimal structure regarding the start and end times of services, significantly reducing the subproblem’s search space. Numerical experiments based on real-life data validate the efficiency of the designed algorithms compared to commercial solvers and highlight the benefits of split services. Sensitivity analysis reveals that as the rental cost function becomes more segmented, more requests are split-serviced, but the times for splitting a request are seldom more than once.

Comparative Analysis of Cooling Load Calculations: CLTD Method vs. Carrier HAP 5.01 Software for Hotel HVAC Design

This study examines the cooling load requirements of a hotel building by comparing two methodologies: the traditional Cooling Load Temperature Difference (CLTD) method and the Carrier Hourly Analysis Program (HAP) 5.01 software. The primary objective is to validate the accuracy and reliability of these methods in calculating cooling loads across different room types, from standard rooms to larger, more complex suites. The results show that the CLTD method consistently yields higher cooling load estimates, with discrepancies ranging from 3% to 14% compared to HAP 5.01 calculations. These differences are most significant in larger rooms, such as suites and owner’s suites, which have more extensive glass areas, higher occupancy, and more heat-generating equipment. The findings indicate that while the CLTD method is valuable for quick, preliminary estimates, the HAP 5.01 software provides a more precise and comprehensive analysis, taking into account hourly variations, equipment schedules, and other factors that impact cooling loads. This research highlights the need for careful selection of the appropriate calculation method to ensure the efficient design of HVAC systems, maximizing energy efficiency, and maintaining occupant comfort. The study concludes that for projects requiring high accuracy, particularly in complex or large spaces, dynamic simulation tools like HAP 5.01 are preferable. Detailed cooling load results and comparisons are provided in the supplementary documentation, offering further insights into the analysis and its implications for HVAC design.

Physical model-assisted deep reinforcement learning for energy management optimization of industrial electric-hydrogen coupling system with hybrid energy storage

Utilizing renewable energy sources (RESs), such as wind and solar, to convert electrical energy into hydrogen energy for industrial users with different types of energy storage can enhance the integration of green electricity, thereby replacing fossil fuels like natural gas. In the energy management optimization of industrial electric‑hydrogen coupling system (EHCS) with high RESs integration, the traditional optimization methods cannot effectively address the nonlinear characteristics of equipment and stochastic RES generation, while the model-free reinforcement learning methods struggle to improve the training efficiency and may lead to constraint violation. To address this challenge, this paper proposes a short time scale energy management approach for EHCS based on physical models assisted deep reinforcement learning (DRL). Firstly, the energy management optimization model of the EHCS is developed based on the operation characteristics of the equipment within the system. Secondly, the energy management model of EHCS is formulated as a DRL framework. The DRL agent is efficiently trained with the assistance of the action transformation which takes into account the physical characteristics of the equipment during operation. Finally, a case study is conducted to verify the effectiveness of the proposed model for achieving high-efficiency agent training, reducing decision-making time for equipment scheduling, addressing the stochastic RES generation, and realizing the economic operation of EHCS with good performance based on the equipment characteristics.

Research on Train Loading and Unloading Mode and Scheduling Optimization in Automated Container Terminals

In some automated container terminals, railway lines have been implemented into the port, saving container transfer time. However, the equipment scheduling level of the railway yard needs to be improved for managers. In the equipment scheduling of loading and unloading containers for railway trains, the operation modes “full unloading and full loading” and “synchronous loading and unloading” are often adopted. Due to the long length of the railway yard and the line of one train, there are two ways to arrange loading and unloading tasks for automated rail-mounted gantry cranes (ARMGs): one is to pre-assign tasks for ARMGs, and the other is to not pre-assign tasks for ARMGs. To investigate the efficacy of these different operation modes and methods of assigning tasks, this study formulated three mixed-integer linear programming (MILP) models with the goal of minimizing the ARMG task completion time. An adaptive large neighborhood search algorithm was used to tackle the scheduling problem. The scheduling effects of different operation modes and methods for assignment tasks were compared in terms of their calculation time and the completion time of ARMG tasks. Notably, the findings reveal that, with an increase in the number of tasks, the “pre-assign” task arrangement had a limited effect on the completion time of the ARMG tasks, made the calculation time shorter, and reduced the complexity of the problem. From the perspective of the completion time of ARMG tasks, the time under the “synchronous loading and unloading” operation mode was less than that of the “full unloading and full loading” operation mode. Therefore, it is recommended that the managers of the railway yard in an automated container terminal adopt the “synchronous loading and unloading” operation mode but determine the task assignment method according to decision time requirements. In addition, when the number of tasks is large, to decrease the time to complete ARMG tasks, the manager can adopt the “non-pre-assign” task distribution method.

Digital twin-driven proactive-reactive scheduling framework for port multi-equipment under a complex uncertain environment

The pervasive uncertainties in multiple port equipment scheduling frequently result in container handling delays or ineffective plans. To address the complexities and uncertainties of port multiple equipment integrated scheduling problem, this paper introduces a Digital Twin-driven (DT-driven) proactive-reactive scheduling framework for the first time. This framework is designed to promptly respond to uncertainties in the scheduling process and provide a transparent visualization of operational information. It specifically tackles the integrated scheduling problem of port quay cranes, Intelligent Guided Vehicles (IGVs), and yard cranes, considering uncertainties such as fluctuations in operating time, equipment failures, and IGV route conflicts. By developing a virtual container port simulation, which features a U-shaped port layout and double-cycling mode drawn from real-world scenarios, the paper evaluates the proposed framework's effectiveness. The experimental results demonstrate that the digital twin framework method significantly improves efficiency and conserves energy. Additionally, in large-scale conditions, the makespan difference between the DT-driven approach and the non-DT-driven approach is as much as 19.56 %. In terms of energy consumption savings, the DT-driven approach's scheduling plan can save 3.67 % of energy consumption under large-scale conditions. Moreover, as the fluctuation index increases, the energy consumption savings become even more significant. This paper also discusses the potential implications of adopting this framework for port companies, highlighting its benefits in enhancing operational and energy efficiency and its incorporation into port management systems. The sensitivity analysis can offer guidance to port companies on optimal equipment allocation strategies.

A systematic approach to manual calibration and validation of building energy simulation

PurposeThis study aims to create an efficient approach to validate building energy simulation models amidst challenges from time-intensive data collection. Emphasizing precision in model calibration through strategic short-term data acquisition, the systematic framework targets critical adjustments using a strategically captured dataset. Leveraging metrics like Mean Bias Error (MBE) and Coefficient of Variation of Root Mean Square Error (CV(RMSE)), this methodology aims to heighten energy efficiency assessment accuracy without lengthy data collection periods.Design/methodology/approachA standalone school and a campus facility were selected as case studies. Field investigations enabled precise energy modeling, emphasizing user-dependent parameters and compliance with standards. Simulation outputs were compared to short-term actual measurements, utilizing MBE and CV(RMSE) metrics, focusing on internal temperature and CO2 levels. Energy bills and consumption data were scrutinized to verify natural gas and electricity usage against uncertain parameters.FindingsDiscrepancies between initial simulations and measurements were observed. Following adjustments, the standalone school 1’s average internal temperature increased from 19.5 °C to 21.3 °C, with MBE and CV(RMSE) aiding validation. Campus facilities exhibited complex variations, addressed by accounting for CO2 levels and occupancy patterns, with similar metrics aiding validation. Revisions in lighting and electrical equipment schedules improved electricity consumption predictions. Verification of natural gas usage and monthly error rate calculations refined the simulation model.Originality/valueThis paper tackles Building Energy Simulation validation challenges due to data scarcity and time constraints. It proposes a strategic, short-term data collection method. It uses MBE and CV(RMSE) metrics for a comprehensive evaluation to ensure reliable energy efficiency predictions without extensive data collection.

Optimizing multiple equipment scheduling for U‐shaped automated container terminals considering loading and unloading operations

AbstractU‐shaped automated container terminals (ACTs) represent a strategic design in port infrastructure that facilitates simultaneous loading and unloading operations. This paper addresses the challenges of scheduling multiple types of equipment, such as dual trolley quay cranes (DTQCs), automated guided vehicles (AGVs), double cantilever rail cranes (DCRCs), and external trucks (ETs) in U‐shaped ACTs. This paper proposes a mixed integer linear programming model for optimizing the multiple equipment scheduling, aiming to minimize container completion time and AGV waiting time simultaneously. This paper customizes a hybrid genetic‐cuckoo optimization algorithm (HGCOA) with double‐point crossover and Lévy flight Cuckoo search strategies. Extensive numerical results show that the proposed HGCOA outperforms the benchmark genetic algorithms in terms of solution quality and computational time while significantly improving efficiency without substantial sacrifices in solution quality compared with the exact solution method. Overall, this study presents a promising solution for enhancing coordination and operation efficiency in U‐shaped ACTs.

Discussion on the Management Strategy of Imaging Medical Equipment under the New Situation

In the realm of healthcare, particularly in the management of medical imaging equipment, unprecedented challenges and opportunities are encountered. This is especially pronounced in the niche domain of gastrointestinal endoscopy, where innovative management strategies are pivotal in the face of evolving circumstances. With the rapid iteration of technology, the escalating scarcity of medical resources, alongside the rising costs of management and maintenance, coupled with the ongoing flux of regulations and standards, these factors collectively constitute the intricate landscape confronting current management of medical imaging equipment. Addressing the pace of technological advancement, employing proactive technological foresight and flexible updating strategies are paramount. This entails not only technological upgrades of the equipment itself but also continuous training and education for operators to ensure their alignment with technological advancements. Given the scarcity of medical resources, enhancing the utilization and efficiency of existing equipment becomes imperative. This may involve optimizing equipment allocation and scheduling, reducing idle time, while also exploring the application of emerging technologies such as telemedicine to expand service coverage. This article will delve into how, amidst new circumstances, effective management strategies can optimize the utilization and maintenance of gastrointestinal endoscopy to cope with challenges such as the rapid pace of technological evolution, scarcity of medical resources, increasing management and maintenance costs, as well as changes in regulations and standards.

An Adaptive Large Neighborhood Search Algorithm for Equipment Scheduling in the Railway Yard of an Automated Container Terminal

In container sea–rail combined transport, the railway yard in an automated container terminal (RYACT) is the link in the whole logistics transportation process, and its operation and scheduling efficiency directly affect the efficiency of logistics. To improve the equipment scheduling efficiency of an RYACT, this study examines the “RYACT–train” cooperative optimization problem in the mode of “unloading before loading” for train containers. A mixed-integer programming model with the objective of minimizing the maximum completion time of automated rail-mounted gantry crane (ARMG) tasks is established. An adaptive large neighborhood search (ALNS) algorithm and random search algorithm (RSA) are designed to solve the abovementioned problem, and the feasibility of the model and algorithm is verified by experiments. At the same time, the target value and calculation time of the model and algorithms are compared. The experimental results show that the model and the proposed algorithms are feasible and can effectively solve the “RYACT–train” cooperative optimization problem. The model only obtains the optimal solution of the “RYACT–train” cooperative scheduling problem with no more than 50 tasks within a limited time, and the ALNS algorithm can solve examples of various scales within a reasonable amount of time. The target value of the ALNS solution is smaller than that of the RSA solution.

Application of Ginni’s mean analysis for identifying the stationary driver enhancing modular construction adoption in the building industry

Modular construction (MC) is recognized as an environmentally friendly and sustainable building method, and it is mandatory for assessing project sustainability. In developed countries, MC is acknowledged as a crucial approach for sustainable development, leading to increased adoption in these nations. However, implementing and incorporating MC in developing nations is a multifaceted task due to infrastructure limitations, economic inequalities, and resource constraints, which limit its widespread adoption. This study aims to find the stationary driver that increase MC usage in developing nations, especially in residential projects. To achieve this objective, a total of 120 building professionals participated in the questionnaire. Following this, the Relative Importance Index and Gini’s Mean analyses were employed to assess the significance of the top 5 ranked drivers and to pinpoint the stationary driver. According to the study results, the five major MC drivers are as follows: Determining the correct weights and dimensions of modules by optimal design; Considering the detailed survey of fabricators’ MC experience and competency; Early screening tests during the selection of fabricators and factories; Early engagement of the heavy lift team in relevant research; and Avoiding suboptimal modular equipment sizes that might restrict transportation options. Additionally, the Gini’s Mean analysis revealed the stationary driver that promotes MC adoption is the need for expert advice for equipment specifications, scheduling, and sequence. The research findings pave the way for the Egyptian construction industry to adopt MC, which will improve the construction sector, particularly residential construction. Furthermore, the results address a knowledge gap about the drivers that enhance the adoption of MC in developing nations.

Understanding the Employment Landscape in People With Systemic Sclerosis.

Systemic sclerosis (SSc) can restrict employment participation. Our objectives were to comparatively evaluate health factors, work factors, and workplace accommodations between those who are employed and those who recently gave up employment. A cross-sectional study was conducted of employed and recently working, but now unemployed, individuals with SSc. Demographics, employment sectors, health factors, flare frequency, work context, and information about the need, availability, and use of workplace supports were collected. Participants were 140 individuals (108 [77.1%] women, 32 [22.9%] men), of whom 110 (78.6%) were employed and 30 (21.4%) were unemployed. Participants worked in education/health/sciences/arts (n = 51, 36.4%), sales/retail (n = 23, 16.5%), banking/insurance/business/technology (n = 22, 15.7%), government (n = 15, 10.7%), construction/utilities (n = 10, 7.1%), and manufacturing/agriculture/mining/logging (n = 10, 7.1%). Employed participants had a lower mean age (48.4 vs 54.3 yrs), and higher level of education (77.3% with postsecondary education vs 22.7% without). Those who had no flares were more frequently employed (41.7%), compared to those who had 1 to 2 flares (35.2%) and ≥ 3 flares (23.1%). The availability of workplace accommodations differed significantly between the employed and unemployed: flexible hours (74.5% vs 40%, P = 0.0005), more rest periods (73.6% vs 46.7%, P = 0.0001), special equipment (82.7% vs 46.7%, P < 0.0001), and work schedule flexibility (66.4% vs 33.3%, P = 0.003). Health factors alone do not differentiate those who are employed and those who gave up employment. This study lays the groundwork for where SSc-specific efforts in workplace policies and practices should be directed, especially workplace support.

Penerapan Metode Waterfall dalam Pengembangan Aplikasi Schedule Maintenance Alat Produksi

PT. Shiroki Indonesia Cikarang is moving on the field of automotive spare parts manufacturing such as window regulator, seat recliner, sear track as well as floor lock. The activity uses a tool/machine as a support to produce spare parts that have good quality. With so many of these machines, then the company needs to do the maintenance and also the maintenance of the machine well. Scheduling of maintenance and maintenance of machinery production equipment still uses the record on board and requires information from the operator to be submitted to the technician. Therefore, it is necessary to create a web-based maintenance schedule application that can facilitate the performance of the staff in making the maintenance schedule of the production equipment more effective and efficient as well as the availability of reports from the maitenance. The research method used is the waterfall method which has five stages namely needs analysis, design, program code making, testing and support or maintenance. The schedule maintenance application was created using the PHP programming language using the CodeIgniter framework and the MySql database.

An advanced control strategy of hybrid cooling system with cold water storage system in data center

The inefficient operation of cooling equipment is a significant impact factor to the high energy consumption of cooling system in data center. This study proposes an advanced model predictive control (MPC) strategy for a hybrid cooling with water storage system to improve energy efficiency and reduce the accumulation of cold storage losses. Mixed integer linear programming (MILP) in MPC strategy is used to optimize the operating parameters under free cooling, hybrid cooling, and mechanical cooling modes, further solving the problem of precise optimization for different modes. Taking Guangzhou city as an example, the equipment scheduling and the appropriate volume of cold water storage tank for MPC strategy are analyzed. The results indicate that, the emergency cold water storage tank 500 m3 only supports the efficient operation of cooling system under the maximum 60 % IT load rate, meanwhile, the optimal tank volume 1400 m3 could meet the 60–100 % IT load rate. Compared to Baseline strategy, the biggest reduction of annual energy consumption using MPC strategy would be attained by 12.19 % under free cooling mode, by 4.04 % under hybrid cooling mode, and by 22.15 % under mechanical cooling conditions at the 55 % IT load rate. Therefore, the MPC strategy proposed in this paper has important guiding significance for energy conservation and carbon reduction in global data centers.

A study on the correlation between physical education courses and students’ health status in colleges and universities based on big data analysis

Abstract Physical education in colleges and universities is of great significance to the development of the whole education chain, to realize the goal of physical education courses in colleges and universities and to promote the physical and mental health development of students, this paper takes the physical education courses and students in Z college and university as the object of research, and proposes a correlation model of the physical education courses in colleges and universities and students’ health status by combining the fuzzy comprehensive evaluation and the Apriori correlation rule algorithm. The weights were calculated after constructing the physical education course quality index system. The quality of physical education courses in college Z was comprehensively evaluated. It was concluded that teaching explanation and demonstration (0.137), venue equipment (0.133) and physical education class schedule (0.113) had the most significant influence on the quality of physical education courses. The quality of physical education courses in the selected colleges and universities was good, scoring 88.93 points. When the quality of physical education courses is excellent and sound, the confidence level of students’ health status corresponding to excellent and sound is greater than 80%, indicating that physical education courses strongly correlate with students’ health. Among the various indicators of courses analyzed using big data, teaching explanation and demonstration, venue equipment, and students’ health status have stronger correlations.

Deep Reinforcement Learning-Based Scheduling for NR-U/WiGig Coexistence in Unlicensed mmWave Bands

This paper investigates the coexistence of the New Radio-based access to Unlicensed spectrum (NR-U) network and the Wireless Gigabit (WiGig) network in unlicensed millimeter-wave (mmWave) bands. To enable the NR-U network to achieve equitable and harmonious spectrum sharing with WiGig systems, we develop two new classes of user equipment (UE) scheduling schemes by exploiting the deep reinforcement learning (DRL) technique. Specifically, we first propose the distributed deep reinforcement learning scheduling (DeepDS) scheme, wherein multiple deep neural networks (DNNs) are used to make decisions for different panels in an independent fashion. Thereafter, to reduce the computational cost of adopting multiple DNNs, we design the centralized deep reinforcement learning scheduling (DeepCS) scheme that introduces the shared DNN framework to perform decisions for all panels in parallel at one time. The objective of both DeepDS and DeepCS is to maximize the total data rate of the NR-U network with as little interference to WiGig systems as possible, while satisfying the quality of service (QoS) requirement for each UE. We first formulate this problem into the constrained Markov decision process framework. To address the multi-constraint issue, we put forth a new DRL algorithm that incorporates the Lagrangian primal-dual optimization into the deep Q-network framework, referred to as adaptive multi-constraint deep Q-network (AMC-DQN). With AMC-DQN, both DeepDS and DeepCS can achieve their goals even without acquiring prior operations about the WiGig network. Simulation results show that compared with the state-of-the-art omniLBT and dirLBT, both DeepDS and DeepCS yield significant performance benefits in terms of the total network data rate. We also demonstrate the ability of DeepDS and DeepCS to satisfy the QoS requirements of different UEs and their robustness against various simulation setups. Furthermore, compared with DeepDS, DeepCS can save a large amount of computational cost although at the expense of a slightly lower data rate.