Batch Scheduling Research Articles

A two-phase approach for the Radiotherapy Scheduling Problem.

The Radiotherapy Scheduling Problem (RTSP) focuses on optimizing the planning of radiotherapy treatment sessions for cancer patients. In this paper, we propose a two-phase approach for the RTSP. In the first phase, radiotherapy sessions are assigned to specific linear accelerators (linacs) and days. The second phase then decides the sequence of patients on each day/linac and the specific appointment times. For the first phase, an Integer Linear Programming (IP) model is proposed and solved using CPLEX. For the second phase, a Mixed Integer Linear Programming (MIP) and a Constraint Programming (CP) model are proposed. The test data is generated based on real data from CHUM, a large cancer center in Montréal, Canada, with an average of 3,500 new patients and 40,000 radiotherapy treatments per year. The results show that in the second phase, CP is better at finding good solutions quickly while MIP is better at closing optimality gaps with more run time. Lastly, a simulation is conducted to evaluate the impact of different scheduling strategies on the outcome of the scheduling. Preliminary results show that batch scheduling reduces patients' waiting time and overdue time.

Batch scheduling with incompatible job families and common due date

This paper explores incompatible batch scheduling with common due date on a single batch processing machine. The problem of minimizing the number of tardy jobs is proved to be NP-hard with respect to id-encoding, and pseudo-polynomial time algorithms for minimizing the number of tardy jobs and the weighted number of tardy jobs are presented, respectively.

When serial batch scheduling involves parallel batching decisions: A branch and price scheme

This paper studies a scheduling problem in which serial and parallel batching decisions must be simultaneously taken. There is a single machine with a fixed capacity that can process multiple jobs at the same time (i.e., parallel batching). Throughout the planning horizon, jobs are released at different instants and they belong to different families. Each job requires one unit of the machine capacity and only jobs of the same family can be processed at the same time. Once a group of jobs is decided to be processed, a preparation time occurs and all jobs are considered completed when the last job of the group finishes its processing (i.e., serial batching). We aim to minimize total flow time (i.e., sum of job completion times, ∑Cj) and total weighted tardiness (∑wjTj) objectives. First, we formulate this problem as an integer linear programming model. Then, we develop a column generation algorithm to solve the linear relaxation which is integrated in a branch and bound tree in search for the integer solution. We use a highly efficient heuristic branching technique for ∑wjTj objective and prove its optimality for ∑Cj objective. Numerical test results show that the algorithm can handle large instances in reasonable computational times.

The integrated orderline batching, batch scheduling, and picker routing problem with multiple pickers: the benefits of splitting customer orders

Fast delivery is one of the most popular services in e-commerce retail. It consists in shipping the items ordered on-line in short times. Customer orders in this segment come with deadlines, and respecting this latter is pivotal to ensure a high service quality. The most time-consuming process in the warehouse is order picking. It consists in regrouping orders into batches, assigning those batches to order pickers, sequencing the batches assigned to each order picker such that the orders deadlines are satisfied, and the picking time is minimized. To speed up the order picking operations, e-commerce warehouses implement new logistical practices. In this paper, we study the impact of splitting the orders (assigning the orderlines of an order to multiple pickers). We thus generalize the integrated orders batching, batch scheduling, and picker routing problem by allowing the orders splitting and propose a route first-schedule second heuristic to solve the problem. In the routing phase, the heuristic divides the orders into clusters and constructs the picking tours that retrieve the orderlines of each cluster using a split-based procedure. In the scheduling phase, the constructed tours are assigned to pickers such that the orders deadlines are satisfied using a constraint programming formulation. On a publicly available benchmark, we compare our results against a state-of-the-art iterated local search algorithm designed for the non-splitting version of the problem. Results show that splitting the customer orders using our algorithm reduces the picking time by 30% on average with a maximum reduction of 60%.

Order batching and picker scheduling in warehouse order picking

This article focuses on the integration of order batching and picker scheduling decisions while taking into account two objectives that have been considered in the literature, namely the minimization of both total travel time to collect all items and makespan of the pickers. This integrated problem not only occurs naturally in wave picking systems in which the latest picking time of orders becomes the key performance metric, but also arises when there is a limit on the picker operating time. We present models that result from combining these objectives and analyze their relationship through bounds. We propose a column generation-based exact algorithm for the integrated problem. The novelty of the proposed approach lies in the ability of efficiently solving the integrated order batching and picker scheduling problem to optimality by designing a column generation subproblem based on the set of batches, which makes it a challenging optimization problem due to its size. We alleviate this difficulty by reformulating this subproblem, which allows efficient implicit enumeration of its variables. We have also devised a Variable Neighborhood Search algorithm used as a subprocedure within the proposed exact solution algorithm. Finally, we conduct experiments on randomly generated instances and show that the proposed algorithms are capable of solving instances with up to 100 orders.

Interactive analysis notebooks on DESY batch resources

Batch scheduling systems are usually designed to maximise fair resource utilisation and efficiency, but are less well designed for demanding interactive processing, which requires fast access to resources while low upstart latency is only of secondary significance for high throughput of high performance computing scheduling systems. The computing clusters at DESY are intended as batch systems for end users to run massive analysis and simulation jobs enabling fast turnaround systems, in particular when processing is expected to feed back to operation of instruments in near real-time. The continuously increasing popularity of Jupyter Notebooks for interactive and online processing made an integration of this technology into the DESY batch systems indispensable. We present here our approach to utilise the HTCondor and SLURM backends to integrate Jupyter Notebook servers and the techniques involved to provide fast access. The chosen approach offers a smooth user experience allowing users to customize resource allocation tailored to their computational requirements. In addition, we outline the differences between the HPC and the HTC implementations and give an overview of the experience of running Jupyter Notebook services.

Robust resource targeting in continuous and batch process

Water is considered a significant resource in process industries. It is essential for planners to target and optimize the use of water as an external resource for industrial operations. Such optimization problems account for uncertainties related to internal resources and must be handled to provide solutions for real plants of industrial relevance. In this paper, these parametric uncertainties are addressed, while targeting resources for continuous and flexible schedule batch process. The proposed robust counterpart formulations include resource minimization constraints for continuous and batch processes to satisfy the demand. Three different robust optimization methodologies are adapted and extended to handle parametric uncertainties associated with internal resources. Assuming bounded and known uncertainty, the resultant formulations are then implemented to literature examples, and the results are compared with the deterministic formulation. The results show that the formulation proposed by Bertsimas and Sim is the most appropriate model for the defined problem because it preserves the linearity and provides a mechanism to control the degree of conservatism, guaranteeing feasibility. The proposed formulations are also explained using illustrative examples estimating the additional 4.04% and 11% requirement of resource in continuous and batch process, respectively, to handle uncertainty with a known risk. This model will assist the planner to decide the resource requirement under uncertain conditions and do the necessary preparation accordingly, and thus, it immunes the process against uncertainties to satisfy demands.Graphic abstract

A batch scheduling model for a three-stage flow shop with job and batch processors considering a sampling inspection to minimize expected total actual flow time

Purpose: This research develops a batch scheduling model for a three-stage flow shop with job processors in the first and second stages and a batch processor in the third stage. The model integrates production process activities and a product inspection activity to minimize the expected total actual flow time.Design/methodology/approach: The problem of batch scheduling for a three-stage flow shop is formulated as a mathematical model, and a heuristic algorithm is proposed to solve the problem. This model applies backward scheduling to accommodate the objective of minimizing the expected total actual flow time.Findings: This research has proposed a batch scheduling model for a three-stage flow shop with job and batch processors to produce multiple items and an algorithm to solve the model. The objective is to minimize total actual time. The resulting production batches can be sequenced between all types of products to minimize idle time, and the batch processor capacity affects the sample size and indirectly affects the production batch size.Originality/value: This research develops a batch scheduling model for a three-stage flow shop constituting job and batch processors and carrying out integrated production and inspection activities to minimize the expected total actual flow time

Optimally Managing Chemical Plant Operations: An Example Oriented by Industry 4.0 Paradigms

Updating industrial facilities to increase the level of automation and digitalization to match Industry 4.0 paradigms has become essential for many companies. Following such a trend, this paper presents a real-time optimization algorithm that plays a central role in a larger project framework devoted to highly interconnecting different network components of an Italian chemical industrial site. The proposed methodology aims at best managing the production rates of various products to fulfill a sales plan organized to satisfy numerous client requests. The considered model takes into account both batch and continuous processes as well as salable and non-storable products. The algorithm structure relies on the use of a non-linear optimization scheme and on the concepts of batch scheduling. Different features of the proposed methodology have been tested on real plant data, showing how the predicted forecast always improved the initial operation plan by considering both aspects of feasibility and economic nature. The use of the proposed algorithm assures the basis for fully integrating the control systems and the selling department of the facility in a more interactive and responsive manner.

Single batch machine scheduling with dual setup times for autoclave molding manufacturing

Batch scheduling involves a machine that can process several jobs simultaneously. The existing literature mainly focuses on the batch-size-dependent setup time. However, setup time depends not only on the batch size but also on technological characteristics. This paper investigates a single batch scheduling problem with dual quantitative and technological setup times extracted from autoclave molding production. A mixed-integer linear programming model (MILP) is established, in which the objective function is to minimize the makespan. The quantitative setup time is linear with the batch size, whereas the technological setup time is nonlinear with the occupied area of all jobs (jobs are represented by rectangles) in a batch. Three lower bounds are presented for evaluating the proposed algorithms. Then, a two-stage approximate algorithm is designed to solve the problem. The first stage concerns the normal batch processing time. In the second stage, an iterative local search method focuses on all terms in the objective function and improves the current solution’s quality. Finally, different scales of instances are designed to test the effectiveness and efficiency of the algorithms. Statistical analysis shows that the proposed algorithm in this paper can handle the problem well compared to other algorithms.

Automated Production Scheduling for Artificial Teeth Manufacturing

In industrial artificial teeth manufacturing, nowadays a high level of automation is utilized to produce a large quantity of teeth in short production cycles. As a large variety of different product shapes and colors have to be processed on a single machine, the creation of efficient production schedules becomes a very challenging task. Due to the complexity of the problem and several cost minimization objectives that need to be considered, there usually is a large potential to improve the currently manually created schedules with automated solution methods. In this paper, we formally specify and solve a novel challenging real-life machine batch scheduling problem from the area of artificial teeth manufacturing. Additionally, we provide a collection of real-life benchmark instances that can be used to evaluate solution methods for the problem. To efficiently solve the problem, we propose an innovative construction heuristic and metaheuristic approach as well as an exact method using constraint programming. An extensive experimental evaluation shows that exact techniques can efficiently solve small scheduling scenarios and can provide optimal solutions for four instances. Furthermore, we show that the proposed metaheuristic approach is able to reach optimal results for small instances and can find high quality solutions also for large real-life benchmark instances.

C2QoS: Network QoS guarantee in vSwitch through CPU-cycle management

Cyber–Physical–Social–System (CPSS) relies on cloud and edge computing for service deployment and acceleration. As an enabling technology of CPSS, network virtualization allows different services to be deployed on a single physical server and provides them with differentiated network performance based on service-level-agreement (SLA). Unfortunately, current virtual switches (vSwitches) cannot guarantee this functionality due to the common resource competition and unpredictable processing capacity. As a software process on the server, vSwitch is allocated limited dedicated CPU cores to implement traffic forwarding for all services. But the QoS strategy inheriting from the hardware switch, does not consider the forwarding tasks’ competition for CPU cores in terms of utilization and timing, thus cannot guarantee the SLA. To solve this critical issue, we propose a CPU-cycle based QoS (C2QoS) strategy to realize the isolation and guarantee of service network performance by managing the usage and scheduling of the IO-dedicated CPU cores in vSwitch. C2QoS includes a CPU-cycle based token bucket mechanism to strictly limit service’s network bandwidth and a hierarchical batch scheduling mechanism to achieve hierarchical latency. Experimental results show that, compared with existing strategies, C2QoS can strictly guarantee the network bandwidth of the services and reduce the service latency by up to 80%.

A Review on Design and Development of Performance Evaluation Model for Bio-Informatics Data Using Hadoop

The paper reviews the usage of the platform Hadoop in applications for systemic bioinformatics. Hadoop offers another system for Structural Bioinformatics to break down broad fractions of the Protein Data Bank that is crucial to high-throughput investigations of (for example) protein-ligand docking, protein-ligand complex clustering, and structural alignment. In specific, we review different applications of high-throughput analyses and their scalability in the literature using Hadoop. In comparison to revising the algorithms, we find that these organisations typically use a realized executable called MapReduce. Scalability demonstrates variable behavior in correlation with other batch schedulers, particularly as immediate examinations are usually not accessible on a similar platform. Direct Hadoop examinations with batch schedulers are missing in the literature, but we note that there is some evidence that the scale of MPI executions is better than Hadoop. The dilemma of the interface and structure of an asset to use Hadoop is a significant obstacle to the utilization of the Hadoop biological framework. This will enhance additional time as Hadoop interfaces, such as enhancing Flash, increasing the use of cloud platforms, and normalized approaches, for example, are taken up by Workflow Languages.

Optimization In a Casing Manufacturing Industry Using Simulation Study

This project describes how Simulation can be applied to optimize a job shop scheduling problem (JSSP) at a precision aerospace components manufacturing company. The product selected for the simulation study is casing. As 8 different casings are manufactured in a single casing manufacturing line, the demand requirements are not met. To identify the bottleneck operations, the existing manufacturing process is simulated using Arena simulation software. The model is simulated for various batch sizes and Scheduling rules and the performance parameters like total casing output, average work in process, average utilization and average flow time were determined. Design of experiments and Interaction plot is used to determine the optimum batch size and scheduling rule. The results showed that for maximum casing output and Resource utilization, the optimum batch size is 5 and optimum scheduling rule is Shortest processing time (SPT) priority rule. To minimize the Average WIP and flow time, the optimum batch size is 2 and optimum scheduling rule is Shortest processing time (SPT) priority rule.

Mathematical modeling using batch scheduling approach for optimal tritium inventory in the ISS of the ITER fuel cycle

An optimal model to schedule fuel regeneration is developed for the Tokamak Cryopumps and Cryogenic Viscous Compressors (CVCs) of the fuel cycle of ITER. The objective is to minimize the tritium inventory inside the Isotope Separation Systems (ISS) while supplying the tokamak’s required amount of fuel considering fueling stages: ramp-up, flat-top, and ramp-down for each tokamak operation mode (Inductive, Hybrid, and Non-Inductive). Case studies of CVC regeneration times are presented to illustrate how CVC regeneration affects the ISS tritium working inventory. Results showed that nine iterations of CVC regeneration are enough to maintain ISS working tritium inventory management during Non-Inductive tokamak operation mode. The developed model uses mixed integer linear programming of the state-task network.

Local search for weighted sum coloring problem

The weighted sum coloring problem (WSCP) is a weighted vertex coloring problem in which a weight is associated with each color. Its aim is to find a coloring of the vertices of a graph with the minimum sum of the costs of the used colors. The WSCP has important applications in the batch scheduling of distributed systems. In this paper, we obtain an integer linear programming (ILP) model for the WSCP, and design an efficient local search algorithm named LS-WSC to solve WSCP. To this end, we propose two ideas to make LS-WSC effective. Firstly, we design an adaptive strategy to change the number of color classes in current solution, which allows LS-WSC to search more feasible solution regions. Secondly, we design a greedy function and a tabu strategy, which make LS-WSC search in right direction. We also design a genetic algorithm GA-WSC to solve WSCP, which is used to experimentally evaluate the result of local search algorithms. The preliminary experimental results show that LS-WSC outperforms the famous CPLEX on almost all instances, and is competitive with the approximate algorithm GREEDY on most benchmark instances with vertex number less than 200. However, the performance of LS-WSC is worse than GREEDY algorithm′s on most benchmark instances with vertex number greater than 200. Overall, the performance of LS-WSC is only marginally better than GA-WSC′s. In order to improve the performance of LS-WSC, we use the configuration checking strategy to update the searching process of LS-WSC, and the new algorithm for solving WSCP is named as CCLS-WSC. The further experimental results show that CCLS-WSC algorithm outperforms the famous CPLEX, the approximate algorithm GREEDY, the genetic algorithm GA-WSC and LS-WSC on all benchmark instances.

A new method for heat integration between nonrepeatable batches

AbstractHeat storage is an important part for heat integration between nonrepeatable batches. This article suggests a new approach in the design of heat storage exchange network for heat integration between nonrepeatable batches. This approach is built on the TRANSSHIPMENT method. The main idea is reusing the waste heat from stream segments that are designed to exchange heat with utilities. These extracted stream segments are added to every upcoming time slice as new energy streams. The possible heat exchange between extracted segments and existing streams in every Time Slice is based on the suggested algorithm which includes the main rules of the Pinch technology. All possible matches are potential heat storage units or subsystems. The efficiency of heat recovery increases if the schedule of nonrepeatable batches is repeated every day. That makes a great impact on the production costs. Otherwise, as it is in the given case study, the waste heat reduction is small and cannot support the investment for the solution. The reduction of hot utility in the presented case study is 43.5% and the reduction of cold utility is 12.1% in the start‐up period of the plant (58.94 hours).

Dynamic Distribution of High-Rate Data Processing from CERN to Remote HPC Data Centers

The prompt reconstruction of the data recorded from the Large Hadron Collider (LHC) detectors has always been addressed by dedicated resources at the CERN Tier-0. Such workloads come in spikes due to the nature of the operation of the accelerator and in special high load occasions experiments have commissioned methods to distribute (spill-over) a fraction of the load to sites outside CERN. The present work demonstrates a new way of supporting the Tier-0 environment by provisioning resources elastically for such spilled-over workflows onto the Piz Daint Supercomputer at CSCS. This is implemented using containers, tuning the existing batch scheduler and reinforcing the scratch file system, while still using standard Grid middleware. ATLAS, CMS and CSCS have jointly run selected prompt data reconstruction on up to several thousand cores on Piz Daint into a shared environment, thereby probing the viability of the CSCS high performance computer site as on demand extension of the CERN Tier-0, which could play a role in addressing the future LHC computing challenges for the high luminosity LHC.

Mid-term prediction of electrical energy consumption for crude oil pipelines using a hybrid algorithm of support vector machine and genetic algorithm

The mid-term electrical energy consumption forecasting for crude oil pipelines is helpful for making important decisions, such as energy consumption target setting, unit commitment, batch scheduling, and equipment monitoring with degraded performance. The electricity energy consumption forecasting during operation is complicated. Therefore, A hybrid prediction method combining genetic algorithm and support vector machine is proposed, which includes four parts: data preprocessing part, optimization part, forecasting part, and evaluation part. The stratified sampling method is adopted to divide the training set and the test set to avoid large deviation caused by sampling stochasticity of small samples. According to the nonlinear relationship between input variable and output variable mapped by SVM technology, genetic algorithm was proposed to optimize the hyperparameters of SVM. For the operation data of three crude oil pipelines in China, the different proportions of data sets are compared and analyzed, the ratio of training set to test set for Pipeline 1, Pipeline 2, and Pipeline 3 is 6:4, 7:3, 8:2, respectively. Comparing the evaluation indicators of GA-SVM with that of five state-of-the-art prediction methods, GA-SVM hybrid model has the best effect in improving the predictive accuracy, and the forecast results are in the best agreement with the actual data.

Research and Implementation of Scheduling Strategy in Kubernetes for Computer Science Laboratory in Universities

How to design efficient scheduling strategy for different environments is a hot topic in cloud computing. In the private cloud of computer science labs in universities, there are several kinds of tasks with different resource requirements, constraints, and lifecycles such as IT infrastructure tasks, course design tasks submitted by undergraduate students, deep learning tasks and and so forth. Taking the actual needs of our laboratory as an instance, these tasks are analyzed, and scheduled respectively by different scheduling strategies. The Batch Scheduler is designed to process tasks in rush time to improve system throughput. Dynamic scheduling algorithm is proposed to tackle long-term lifecycle tasks such as deep learning tasks which are hungry for GPU resources and have dynamically changing priorities. Experiments show that the scheduling strategies proposed in this paper improve resource utilization and efficiency.