Predictive Scheduling Research Articles

A coordinated predictive scheduling and real-time adaptive control for integrated building energy systems with hybrid storage and rooftop PV

A coordinated predictive scheduling and real-time adaptive control for integrated building energy systems with hybrid storage and rooftop PV

تحسين خوارزمية الجولة الدائرية باستخدام نماذج التعلم الآلي (KNN) في جدولة وحدة المعالجة المركزية

Round Robin is considered as one of the most practically recognized process scheduling algorithms in CPU scheduling because it is simple and fair. However, the efficiency of Round Robin depends a lot upon the selection of an optimal time quantum. If the quantum is too small, then frequent context switches are needed by the CPU; therefore, the overhead increases, and thus, the average waiting time for the processes also becomes long. As a result, system performance is reduced as more and more CPU time is used up in context switching, instead of task execution. it may behave similarly to the First Come First Serve (FCFS) algorithm if the time quantum is excessively long, which leads to extended average waiting time. This paper proposes an improved Round Robin algorithm by incorporating machine learning, which optimally determines the time quantum dynamically. More precisely, the K-Nearest Neighbors algorithm will be used, with NumPy in charge of data processing, for the runtime prediction of an optimal time quantum considering characteristics of processes. The experimental results showed a considerable improvement in the parameters of average waiting, turnaround time, and the number of context switches with respect to the traditional Round Robin algorithm. Results indicated that machine learning efficiently modifies the predictable scheduling algorithms to make the scheduling process adaptive and efficient in operating systems. Key words: dynamic round robin, classical round robin, burst time, machine learning.

EvPowerStop - Electric Vehicle Station Finder and Slot Scheduler

With the increasing adoption of electric vehicles (EVs) worldwide, a significant challenge has emerged for EV drivers: locating compatible charging stations in real-time and securing slots for efficient use. The transition to electric mobility has created a demand for seamless infrastructure that supports the dynamic needs of EV drivers. Currently, the lack of centralized information on charging station availability and compatibility with specific vehicles leads to inconveniences and delays, reducing the overall efficiency of electric vehicle usage. This project addresses this gap by developing an EV station finder and slot scheduling application—EV Power Stop—designed to simplify the charging process and ensure that EV drivers can plan their trips with confidence. The broader context of this work lies in the realm of smart cities and sustainable transportation. As nations push toward carbon-neutral goals, the role of electric vehicles becomes crucial in reducing emissions and lowering dependency on fossil fuels. For this shift to be effective, smart solutions that integrate transportation with infrastructure are vital. This project aligns with the goals of smart urban mobility by offering a system that leverages real-time data to provide EV drivers with the most up-to-date information on charging station availability. The proposed system uses Flutter as a cross-platform framework for the frontend and Firebase as the backend for managing user data, station information, and slot booking in real-time. APIs like Google Maps are integrated to offer navigational support, ensuring users can easily find and access nearby stations. Key features include user authentication, slot booking, real-time updates on availability, and route guidance. The system is designed with scalability, performance, and security in mind, ensuring that it can handle a growing user base while maintaining quick response times. Initial testing shows that the app responds within 2 seconds for most user interactions, meets usability standards, and securely manages personal data. Future developments will focus on integrating additional functionalities like iOS support, predictive scheduling, and automated slot management to improve efficiency further. This project demonstrates the potential for integrating real-time systems with urban mobility to enhance the electric vehicle user experience and support broader sustainability goals..

A-195 Combination of Lean Mindset, Robotics and Digitalization allowed us to manage a 2.5 times increase in workload in our central reference lab

Abstract Background Our Agilus Diagnostics, Gurugram, India is a Central Reference Laboratory and part of the pan-India network over 400 lab in India. Our Gurugram lab delivers 3,200 tubes in the serum work area in 17 hour operations daily and is ISO 15189 and CAP accredited.In Jan 2023, our lab started to support Govt PHC Program samples and our workload increased to 8000 tubes. This was a clear operational and clinical challenge for us with a 2.5 x jump in workload. To achieve this goal, a lean mindset with process improvement steps were to be applied required across the total testing process. This lean approach was supported by cobas ® Lab Automation systems managed by navify ® Lab Operations informatics automation solution. This poster explains our journey of planning, implementing, and delivering on our goal of predictable operational efficiency when faced with this challenge. Methods We took 4 major steps We redesigned lab processes using lean process improvement operational dashboards. This poster will show the metric we track which help eliminate waste in the system. We optimized our total lab automation solution with smart load balancing across cobas ® Pro integrated analyzers and staff friendly recursive workflow. Clear and strict use of off peak hours for maintenance and QC procedures and standardized predictable schedule of operation. This poster will share our time table with key milestones across the 24 hours We built on the Informatics solution using 5 layered Auto verification rules based decision framework. IQC, Systems Flags check, Sample Quality checks, Biological Reference check, custom Auto verification Range, Critical ranges Delta check Criteria, Clinical Correlation Rule are all part of the auto verification logic framework. Results We were able to deliver the following results.Lean process design led to 53% simplification. Number of process steps reduced to 29 from 62 steps across the total testing process. Total Lab Automation has helped improve 90 percentile production TAT by 46% from 250 mins to 135 mins Our 90 % TAT CV has significantly reduced making our production predictable. Over 17,500 results / day reported are subject to auto verification with 87 % of all results auto-verified and auto-approved. Not a single additional employee was added in the analytical or post analytical areas of the lab operations Conclusions Lean Mindset complemented by Total Laboratory Automation along with Decision Automation on an Informatics solution has enhanced lab performance in terms of TAT, Efficiency, and Staff Motivation. This poster will also demonstrate using incremental effectiveness analysis comparative data of key performance indicators, such predictability in production, and efficiency of workforce, senior staff time saved, before and after technology deployment. The challenge of increase in sample workload is real in India and this approach allows for a sustainable way to ensure patient impact with minimum burden on our workforce.

Spectrum: Speedy and Strictly-Deterministic Smart Contract Transactions for Blockchain Ledgers

Today, blockchain ledgers utilize concurrent deterministic execution schemes to scale up. However, ordering fairness is not preserved in these schemes: although they ensure all replicas achieve the same serial order, this order does not always align with the fair, consensus-established order when executing smart contracts with runtime-determined accesses. To preserve ordering fairness, an intuitive method is to concurrently execute transactions and re-execute any order-violating ones. This in turn increases unforeseen conflicts, leading to scaling bottlenecks caused by numerous costly aborts under contention. To address these issues, we propose Spectrum, a novel deterministic execution scheme for smart contract execution on blockchain ledgers. Spectrum preserves the consensus-established serial order (so-called strict determinism) with high performance. Specifically, we leverage a speculative deterministic concurrency control to execute transactions in speculation and enforce an agreed-upon serial order by aborting and re-executing any mis-speculated ones. To overcome the scaling bottleneck, we present two key optimizations based on speculative processing: operation-level rollback and predictive scheduling, for reducing both the overhead and the number of mis-speculations. We evaluate Spectrum by executing EVM-based smart contracts on popular benchmarks, showing that it realizes fair smart contract execution by preserving ordering fairness and outperforms competitive schemes in contended workloads by 1.4x to 4.1x.

A multipath scheduler based on cross-layer information for low-delay applications in 5G edge networks

Virtual Reality (VR) applications that require extremely low delay and high image quality are widely used in online games and other 5G scenarios, becoming a hot research field in recent years. However, the limited bandwidth in 5G edge networks fails to meet the peak rate requirements for multiple VR flows. MPTCP is suitable for 5G edge networks, supporting the simultaneous use of multiple networks on mobile devices. Nevertheless, accurately scheduling VR data blocks to different subflows to satisfy their low delay requirements is challenging due to their microburst characteristic. In this paper, we propose a novel MPTCP scheduler for cloud VR applications in 5G edge networks, called the Cross-layer information-based one-way delay Predictive Scheduler (CPS). CPS accurately predicts one-way delay by incorporating cross-layer information from both the application and edge wireless sides, and adaptively schedules VR data blocks to the optimal subflow. Experimental results show that CPS outperforms existing strategies, supporting 125% more users for VR applications in the typical scenario. CPS successfully meets the quality of experience needs of more users, providing a promising solution for large-scale deployment of cloud VR services in 5G edge networks.

Distributed robust scheduling optimization for energy system of steel industry considering prediction uncertainties

Predictive scheduling is commonly deployed for the energy systems in the steel industry, while the uncertainties caused by the predictions can lead to under-optimization or over-adjustment. In order to solve this problem, a novel distributed robust optimization framework is proposed in this study. A Robust Optimization (RO) model is established at first to mathematically address the prediction uncertainties, where the gas, steam, and electricity networks are considered as distributed participants, with their operating costs and CO2 emissions costs as the objectives for minimization. Then the vanilla 2-block Alternating Direction Multiplier Method (ADMM) is extended to a 3-block structure in this study for solving the established RO model in a distributed manner. Considering the superiority on efficiency, the Column-and-Constraint Generation (C&CG) algorithm is deployed for each block. To ensure convergence considering the extensive interaction between the updated dual and coupling variables in each iteration of the overall ADMM and each C&CG, an adaptive bridging method is developed in this study. Finally, experiments using real data from a steel plant in China demonstrate that the proposed framework has a deviation of less than 1.33 % while with higher efficiency and better robustness compared to other algorithms.

A novel predictive–reactive scheduling method for parallel batch processor lot-sizing and scheduling with sequence-dependent setup time

In practical printed circuit boards (PCB) drilling production systems, multiple spindle processing capabilities and uncertainties are commonplace. This paper addresses predictive–reactive scheduling, considering parallel batch processor lot-sizing and scheduling with sequence-dependent setup times in a dynamic environment. We propose a predictive scheduling algorithm named S-AGAIG to balance machine spindle utilization and order tardiness in steps. Multiple splitting solutions are first formed by using a split heuristic algorithm to size the sub-lots of orders. Then the adaptive genetic algorithm with iterated greedy search is applied to select the solutions and scheduling. Furthermore, we present a schedule repair strategy based on sub-lot co-processing considering the impact of critical sub-lots, and construct a scheduling stability metric for rescheduling. In 36 sets of case experiments encompassing diverse load and machine type configurations, S-AGAIG showcases a 21.27% enhancement in average tardiness performance when compared to its closest rival. Across 28 cases involving varying disturbances, the framework demonstrates exceptional robustness.

Computational-Intelligence-Based Scheduling with Edge Computing in Cyber-Physical Production Systems.

Real-time performance and reliability are two critical indicators in cyber-physical production systems (CPPS). To meet strict requirements in terms of these indicators, it is necessary to solve complex job-shop scheduling problems (JSPs) and reserve considerable redundant resources for unexpected jobs before production. However, traditional job-shop methods are difficult to apply under dynamic conditions due to the uncertain time cost of transmission and computation. Edge computing offers an efficient solution to this issue. By deploying edge servers around the equipment, smart factories can achieve localized decisions based on computational intelligence (CI) methods offloaded from the cloud. Most works on edge computing have studied task offloading and dispatching scheduling based on CI. However, few of the existing methods can be used for behavior-level control due to the corresponding requirements for ultralow latency (10 ms) and ultrahigh reliability (99.9999% in wireless transmission), especially when unexpected computing jobs arise. Therefore, this paper proposes a dynamic resource prediction scheduling (DRPS) method based on CI to achieve real-time localized behavior-level control. The proposed DRPS method primarily focuses on the schedulability of unexpected computing jobs, and its core ideas are (1) to predict job arrival times based on a backpropagation neural network and (2) to perform real-time migration in the form of human-computer interaction based on the results of resource analysis. An experimental comparison with existing schemes shows that our DRPS method improves the acceptance ratio by 25.9% compared to the earliest deadline first scheme.

Eco4cast: Bridging Predictive Scheduling and Cloud Computing for Reduction of Carbon Emissions for ML Models Training

eco4cast: Bridging Predictive Scheduling and Cloud Computing for Reduction of Carbon Emissions for ML Models Training

A Dynamic Staffing & Scheduling Solution: The Build and Implementation of a Logistics Engine to Optimize Nurse Schedules and Rosters

The use of electronic scheduling and staffing tools should be standard practice in health care. Gone are the days of paper scheduling, replaced with mobile apps and sleek desktop applications. The frontline clinician has access to scheduling, paid time off (PTO), trades, and timecard management in the palm of their hand. Ease of use and work flexibility are top of mind for many frontline clinicians and leaders. Yet even with the newest technology and sleekest design, there continues to be a gap in functionality and practice: predictive scheduling. The tools facilitate the management of actions such as balancing a schedule, automating trades, simple approval of time off, and clean record keeping, but have limited decision support functionality to accurately predict how many nurses should be scheduled to match projected demand, or how many nurses should be hired to meet the schedules modeled to match that projected demand. Therefore, nurse managers and leaders spend hours trying to figure out how many nurses are needed on a schedule that would require as little rebalancing as possible for the actual shift in hopes of limiting the chance of cancelations or needing to expensively recruit for unfilled shifts with incentive pay or overtime. The need for decision support tools for predictive scheduling is paramount and critical for effective labor management. To overcome the gap in technology and capability, St. Luke’s Health System designed, built, and implemented their own logistics engine to move from reactive nurse staffing to proactive nurse scheduling, reaping the rewards of managerial time saved, controlled labor cost, a more predictable roster, and a more satisfied nursing workforce.

Care workers’ sense of responsibility, working conditions, and the quality of care in South Korea

As countries like South Korea expand their paid care services, ensuring quality care has become critical. Since care work involves significant emotional labor, a worker’s sense of responsibility for the care recipient's well-being affects the quality of care delivered. In this study, we explore this particular determinant of quality care that has been underexplored to better understand its nature. However, a worker's sense of responsibility or commitment level is not static and varies depending on various factors including working conditions. Using 2018 Korean childcare and eldercare survey data, we empirically examine the relationship between a worker's commitment levels and working conditions by conducting Tobit and generalized maximum entropy (GME) analyses. Results indicate that training, shorter commutes, predictable schedules, and easy interactions with the care recipient’s family are associated with higher levels of commitment. Our findings highlight the importance of supportive working conditions in promoting quality care.

URLM: Utilising reinforcement learning to schedule subflows in MPTCP

SummaryParallel transmission in multiple access networks using MultiPath TCP (MPTCP) greatly enhances the throughput. However, critical packet disorder is commonly observed due to traffic fluctuation and path diversity. Although several predictive scheduling algorithms have been proposed to solve this problem, they cannot accommodate prediction accuracy and real‐time adaptation simultaneously in a dynamic network environment. The time overhead in modifying scheduling parameters to adapt to network changes leads to performance degradation in throughput and packet disorder. In this study, we propose a scheduling algorithm called Utilising Reinforcement Learning to Schedule Subflows in MPTCP (URLM). We apply reinforcement learning to select an optimal scheduling parameter in real time, which brings significant time benefits for modifying the parameters. The simulation comparison experiments show that URLM reduces the average number of out‐of‐order packets and the time overhead in adapting to network changes while improving global throughput.

An inventory data-driven model for predictive-reactive production scheduling

Scheduling is a complex task due to the need to optimise multiple competing objectives and react to unpredictable events that may occur during production execution. The strategy of predictive-reactive scheduling can be used to reconcile the conflict between the original schedule and the current shop floor situation. This study seeks to present an inventory data-driven predictive-reactive production scheduling model that supports the evolving concepts of the Industry 4.0. Periodically, a machine learning technique provides predictive scheduling considering a best-case scenario according to an established Key Performance Indicator (KPI). Then, material non-availability causes disruptions in production, which triggers the Simulation-Based Optimization (SBO) method to handle these events. Thus, SBO provides a reactive schedule with the best set of priority rules to sequence jobs on each machine according to the data on the shop floor. This model was validated with a real case study using data collected from a metal-mechanical company. Considering the service level KPI, the results showed that the model is able to find a better solution in the compared scenarios. Therefore, even in a dynamic and stochastic scenario, with machine breakdowns, quality problems, raw material delays, and accuracy issues, the model proved efficient in mitigating these variations’ effects.

Thermal and energy performance evaluation of a full-scale test cabin equipped with PCM embedded radiant chilled ceiling

The escalating global demand for space cooling has led to the emergence of new cooling technologies, including the phase change material embedded radiant chilled ceiling (PCM-RCC) system. This technology improves energy efficiency and indoor environmental quality, while also offering demand-side flexibility. The present study experimentally evaluates the thermal efficiency and energy performance of a PCM-RCC system in a full-scale test cabin equipped with PCM panels. Here, the transient thermal behaviour of PCM ceiling panels besides the cooling energy delivered during charging-discharging cycles are examined. The indoor thermal comfort and peak electricity demand reduction enabled by the present PCM-RCC are also discussed. The results reveal that chilled water circulation for 4–5 h overnight was sufficient to fully recharge the PCM panels. Over 80% of the occupancy time was classified as “Class B″ thermal comfort according to ISO 7730. The system's daily electricity usage was mostly concentrated during off-peak hours, accounting for ∼70% of the total usage. While the controlling schedule used in this study responded to the transient thermal behaviour of the indoor space and PCM ceiling panels, a more dynamic, predictive schedule is necessary to improve the system's overall efficiency and further enhance indoor thermal comfort in response to the changing environmental conditions.

Robustness Prediction in Dynamic Production Processes—A New Surrogate Measure Based on Regression Machine Learning

This feasibility study utilized regression models to predict makespan robustness in dynamic production processes with uncertain processing times. Previous methods for robustness determination were computationally intensive (Monte Carlo experiments) or inaccurate (surrogate measures). However, calculating robustness efficiently is crucial for field-synchronous scheduling techniques. Regression models with multiple input features considering uncertain processing times on the critical path outperform traditional surrogate measures. Well-trained regression models internalize the behavior of a dynamic simulation and can quickly predict accurate robustness (correlation: r>0.98). The proposed method was successfully applied to a permutation flow shop scheduling problem, balancing makespan and robustness. Integrating regression models into a metaheuristic model, schedules could be generated that have a similar quality to using Monte Carlo experiments. These results suggest that employing machine learning techniques for robustness prediction could be a promising and efficient alternative to traditional approaches. This work is an addition to our previous extensive study about creating robust stable schedules based on deep reinforcement learning and is part of the applied research project, Predictive Scheduling.

Hybrid Modeling and Real-time Predictive Scheduling of Wet Flue Gas Desulfurization for Energy Saving and Life Extension

Sulfur dioxide (SO2) is one of the most important air pollutants emitted by power plants and ironworks. To effectively remove SO2 from flue gas, wet flue gas desulfurization (WFGD) technology is widely adopted by most of China’s coal-fired power plants. However, one of the most important issues of WFGD is the relatively high energy consumption, with the main contribution from the power consumed by circulating pumps in the absorber. In this paper, hybrid modeling and real-time predictive scheduling of a wet flue gas desulfurization (WFGD) system for energy saving and life extension are proposed. The hybrid model is based on the mechanism of the SO2 absorption process and modified based on the operation data using a particle swarm optimization (PSO) algorithm. The prediction model can describe the SO2 absorption process well, and the root-mean-square error (RMSE) of the model is 2.20 mg/m3. Effects of parameters such as the inlet SO2 concentration, pH, flue gas flow rate, and combinations of circulating pumps on the outlet SO2 concentration are further explored. The optimal combination of circulating pumps under different conditions is calculated. Finally, a real-time predictive scheduling strategy is proposed and evaluated under different parameter settings. The system shows the best performance when the upper limit, the target of outlet SO2 concentration, and the lower limit are set to 30, 15, and 10 mg/m3, respectively. The energy consumption of the system is 2564.6 kW, which is 10.6% lower than that before optimization.

Urban Taxi Slogans: The People's Arts

Urban Taxi Slogans: The People's Arts

A Parallel Fast-Track Service Restoration Strategy Relying on Sectionalized Interdependent Power-Gas Distribution Systems

In the distribution networks, catastrophic events especially those caused by natural disasters can result in extensive damage that ordinarily needs a wide range of components to be repaired for keeping the lights on. Since the recovery of system is not technically feasible before making compulsory repairs, the predictive scheduling of available repair crews and black start resources not only minimizes the customer downtime but also speeds up the restoration process. To do so, this article proposes a novel three-stage buildup restoration planning strategy to combine and coordinate repair crew dispatch problem for the interdependent power and natural gas systems with the primary objective of resiliency enhancement. In the proposed model, the system is sectionalized into autonomous subsystems (i.e., microgrid) with multiple energy resources, and then concurrently restored in parallel considering cold load pick-up conditions. Besides, topology refurbishment and intentional microgrid islanding along with energy storages are applied as remedial actions to further improve the resilience of interdependent systems while unpredicted uncertainties are addressed through stochastic/information gap decision theory (IGDT) method. The theoretical and practical implications of the proposed framework push the research frontier of distribution restoration schemes, while its flexibility and generality support application to various extreme weather incidents.

Machine-Learning Approach to Predict Total Fabrication Duration of Industrial Pipe Spools

In industrial modular construction, components fabricated in the shop are assembled into modules before being transported to construction sites. This approach offers improved construction efficiency, higher quality products, and the opportunity to work in a controlled environment that offers more predictable schedules. A distinct category of modularized construction is that of piping structures where prefabricated segments of pipes, called pipe spools, are arranged and attached to a steel structure to speed up on-site assembly. Determining when a particular pipe spool is ready for delivery to the site is affected by many factors, including material availability, workforce performance and availability, shop loading conditions, and shop capacity. These factors combine to produce a high degree of uncertainty when estimating the fabrication duration. This paper investigates the improvement of delivery time prediction of pipe spools that are fabricated in a manufacturing facility. The study’s main contribution is to propose a data-driven machine-learning model that accurately predicts the total fabrication duration in a pipe spool fabrication facility. The developed model is expected to assist production managers to plan for the appropriate workforce and material delivery requirements. The proposed approach utilizes a combination of product composition and real-time loading status of the shop to develop and examine several tree-based machine-learning models. The approach was evaluated using records of 6,989 pipe spools. The overall prediction performance was improved by a reduction of 35% in the mean absolute percentage error compared to original heuristic estimates. The results also show a strong positive correlation of 0.8 between the actual and predicted values.