Two-level Scheduling Approach Research Articles

Application of Polling Scheduling in Mobile Edge Computing

With the Internet of Things (IoT) development, there is an increasing demand for multi-service scheduling for Mobile Edge Computing (MEC). We propose using polling for scheduling in edge computing to accommodate multi-service scheduling methods better. Given the complexity of asymmetric polling systems, we have used an information-theoretic approach to analyse the model. Firstly, we propose an asymmetric two-level scheduling approach with priority based on a polling scheduling approach. Secondly, the mathematical model of the system in the continuous time state is established by using the embedded Markov chain theory and the probability-generating function. By solving for the probability-generating function’s first-order partial and second-order partial derivatives, we calculate the exact expressions of the average queue length, the average polling period, and the average delay with an approximate analysis of periodic query way. Finally, we design a simulation experiment to verify that our derived parameters are correct. Our proposed model can better differentiate priorities in MEC scheduling and meet the needs of IoT multi-service scheduling.

Research on a new two-level scheduling approach for unmanned surface vehicles transportation containers in automated terminals

This paper investigates a two-level scheduling method to improve the operational efficiency of automated terminals by utilizing unmanned surface vehicles (USVs) to transport containers between different berths (exclusively for USVs) in the seaport. To acquire efficient and conflict-free USVs transportation schemes, we established the upper-level mission planning model to minimize the total USVs operation costs by considering the number of USVs deployed, carrying capacity and berthing time intervals, etc. Then, the lower-level collaboration control model is developed to design the predictive controller by considering the rudder and motion constraints, which can reduce path tracking errors for USVs sailing. Then, the chaotic quantum-behaved particle swarm algorithm (cQPSO) and the multi-population genetic algorithm (mGA) are combined to construct a new hybrid algorithm (cQPSO-mGA) that can effectively solve the two-level scheduling model. Finally, experimental results prove that the proposed approach can acquire a low-cost transportation scheme and satisfactory path tracking results for USVs scheduling. Moreover, compared with the selected well-known algorithms, the cQPSO-mGA has better solving accuracy and calculating speed for our problem.

Achieving Fairness-Aware Two-Level Scheduling for Heterogeneous Distributed Systems

In a heterogeneous distributed system composed of various types of computing platforms such as supercomputers, grids, and clouds, a two-level scheduling approach can be used to effectively distribute resources of the platforms to users in the first-level, and map tasks of the users in nodes for each platform in the second-level for executing many-task applications. When scheduling heterogeneous resources, service providers of the system should consider the fairness among multiple users as well as the system efficiency. However, the fairness cannot be achieved by simply distributing an equal amount of resources from each platform to every user. In this paper, we investigate how to address the fairness issue among multiple users in a heterogeneous distributed system. We present three first-level resource allocation policies of a provider affinity first policy, an application affinity first policy, and a platform affinity based round-robin policy, and two second-level task mapping policies of a most affected first policy and a co-runner affinity based round-robin policy. Using trace-based simulations, we evaluate the performance of various combinations of the first and second level scheduling policies. Our extensive simulation results demonstrate that the first-level policy plays a crucial role to achieve relatively good fairness.

Scheduling Optimization of Time-Triggered Cyber-Physical Systems Based on Fuzzy-Controlled QPSO and SMT Solver

The time-triggered communication paradigm is a cost-efficient way to meet the real-time requirements of cyber-physical systems. It is a non-deterministic polynomial NP-complete problem for multi-hop networks and non-strictly periodic traffic. A two-level scheduling approach is proposed to simplify the complexity during optimization. In the first level, a fuzzy-controlled quantum-behaved particle swarm optimization (FQPSO) algorithm is proposed to optimize the scheduling performance by assigning time-triggered frame instances to the basic periods of each link. In order to prevent population from high aggregation, a random mutation mechanism is used to disturb particles at the aggregation point and enhance the diversity at later stages. Fuzzy logic is introduced and well designed to realize a dynamic adaptive adjustment of the contraction–expansion coefficient and mutation rate in FQPSO. In the second level, we use an improved Satisfiability Modulo Theories (SMT) scheduling algorithm to solve the collision-free and temporal constraints. A schedulability ranking method is proposed to accelerate the computation of the SMT-based incremental scheduler. Our approach can co-optimize the jitter and load balance of communication for an off-line schedule. The experiments show that the proposed approach can improve the performance of the scheduling table, reduce the optimization time, and reserve space for incremental messages.

Migrating the fair share algorithm from a distribution to a production planning environment

This paper investigates the ability to migrate the fair share algorithm from a distribution to a production planning environment. In a semi-process-based production system, such as that of the photographic film producer Agfa, the availability of the intermediate product is then the limiting constraint steering the fair share algorithm for the end-product lotsizing decision process. The manufacturing model of Agfa is typically semi-process, where a first stage produces a limited number of intermediate products. The second stage is flow oriented and converts the intermediates into many distinct end-products. The planning method currently implemented within Agfa is a two-level scheduling approach. First, it establishes a cyclical volume plan at the intermediate product level, which is then used as an input constraint for the secondary problem of determining end-product lotsizes. As an alternative to the traditional model, where the end-product lotsizes are determined based on the standard EOQ formula, this investigation suggests the end-product mix decisions to be governed by a tuned fair share algorithm. The paper discusses this algorithm with its parameter settings, the impact on stock values, on service levels, and on set-up and inventory holding costs. The results of both algorithms are compared. This investigation proves that the combination of a cyclical volume plan, at the intermediate product level, combined with fair share mix decisions for the end-product lotsizes, delivers the needed service level with lower inventory levels and reduced operational costs. The main benefit of the model integrating volume planning and mix decisions is its ability to reduce demand amplifications, prohibiting market demand nervousness (amplified by the Forrester effect) from entering into upstream operations. The reduced nervousness allows a major reduction in needed safety stock at the intermediate product level.

Scheduling printed circuit board production systems using the two-level scheduling approach

Abstract In a multiple flowline manufacturing system, such as a typical printed circuit board (PCB) production system, scheduling tasks include determining due dates of customer orders, assigning products to daily production, and releasing boards to production lines each day. The main challenge for the scheduling system is to coordinate different production resources so they can work together to improve system performance. The Bi-Level Intelligent Scheduling System (BLISS) was developed, using a two-level scheduling approach integrating product-level and board-level scheduling, to schedule jobs for PCB manufacturing systems under a daily through weekly planning horizon. A utility function is added to determine lot size, and four sequencing rules are evaluated to find the most suitable rule for the scheduling schemes under study. The lot sizing and sequencing decisions are integrated with the simulation module to generate more accurate schedules, with special emphasis put on the need for coordination. Performance of the two-level scheduling approach is evaluated and compared with that of using product-level or board-level scheduling alone under a variety of manufacturing scenarios. The performance is analyzed under various environments such as different degrees of due date tightness, order variation, and product variation. Results show the strength of the two-level scheduling approach, especially in “mission-critical” situations of tight due dates, high variation of orders, and highly changeable products. The findings also provide guidelines for applying those scheduling schemes effectively and efficiently.

Scheduling printed circuit board production systems using the two-level scheduling approach

In this paper, the development of multiple-input multiple-output (MIMO) subnets is discussed. Hierarchical time-extended Petri nets (H-EPNs), a form of extended Petri nets, allow the development of structured MIMO subnets to model complex system functionalities. In addition, H-EPNs also provide a means to convert such a MIMO structure to a single-input single-output (SISO) net that can easily be integrated within a top-down system decomposition. The activator arc definition is instrumental in deriving this transformation. The activator arc definition and the H-EPN subnet definitions are integrated with the SPNP (stochastic Petri net package) for simulation and analysis purposes. The activator arc extension allows a truly hybrid approach to Petri net (PN)-based systems modeling, analysis, and development of complex PN models of automated manufacturing systems. The case study emphasizes the usefulness of this extension in studying important issues such as static priority scheduling, dynamic failure recognition, and rescheduling in manufacturing systems. The SPNP package, suitably modified to handle the H-EPN extensions, is used for analysis and verification.