Persistent Unpredictability: Analyzing Experiences with the First Statewide Scheduling Legislation in Oregon

Based on 98 in-depth interviews with workers and managers, the authors analyze the effectiveness of Oregon’s Fair Workweek Act, the first statewide scheduling legislation. Overall, findings show limited evidence of the law’s efficacy to improve workers’ schedules. The authors discuss three factors that are likely to explain this shortcoming: lack of adequate funding for education about the law and for enforcement, the inclusion of provisions that undermine the intent of fair scheduling legislation, and the ability of employers to interpret the law with substantial leeway. In this context, the authors consider the persistence of unpredictable scheduling practices a form of “flexible discipline,” even under Fair Workweek legislation. This article contributes to the literature on unpredictable scheduling by showing that in order to address this problem, legislation must include robust funding for education, implementation, and enforcement and must avoid options for workers to waive their rights to predictability pay, which as part of the act is intended to compensate employees for last-minute schedule changes.

An Improved Proportional Fair Scheduling Algorithm for Downlink LTE Cellular Network

Network providers of LTE networks can achieve maximum gain and Quality of Service (QoS) requirement of their users by employing a radio resource management technique that has the ability to allocate resource blocks to users in a fair manner without compromising the capacity of the network. This implies that for a better performing LTE network, a fair scheduling and balanced QoS delivery for various forms of traffic are needed. In this paper an improved proportional fair scheduling algorithm for downlink LTE cellular network has been developed. This algorithm was implemented using a MATLAB-based System Level simulator by Vienna University. The developed algorithm was compared to other scheduling algorithms such as the Proportional Fair (PF) algorithm, Best Channel Quality Indicator (CQI), and Round Robin (RR) scheduling methods. The system performance was also analyzed under different scenarios using different performance metrics. The achieved results showed that the developed algorithm had a better throughput performance than the Round Robin and Proportional fair scheduling. The developed algorithm shows improved cell edge throughputs of about 19.2% (as at 20 users) and 9.1% higher for cell edge users without and with mobility impact respectively. The Best CQI algorithm had higher peak throughput values but the fairness was highly compromised. The developed algorithm outperforms the Best CQI by 136.6% without the impact of mobility. Finally, in dense conditions, the developed algorithm still outperforms the other algorithms with a QoS metric of 4.6% increment when compared to the PF algorithm which was the closest competitor. Keywords: UE, eNodeB, Scheduling, Proportional Fair, LTE,

A Multi-Tier Architecture for the Management of Supply Chain of Cloud Resources in a Virtualized Cloud Environment

As cloud resource demand grows, supply chain management (SCM), which is the core function of cloud computing, faces serious challenges. Quite a number of techniques have been proposed by many researchers for such a challenge. As such, numerous proposed strategies are still under reckoning and modification so as to enhance its potential. An optimized dynamic scheme that combined several algorithms' characteristics was proposed to map out such a challenge. The hybridized proposed scheme involved the meta-heuristic swarm mechanism of ant colony optimization (ACO) and deterministic spanning tree (SPT) algorithm as it obtained faster convergence chain, ensured resource utilization in least time and cost. Extensive experiments conducted in cloudsim simulator provided an efficient result in terms of minimized makespan time and throughput as compared to SPT, round robin (RR), and pre-emptive fair scheduling algorithm (PFSA) as it significantly improves performance.

IMPLEMENTASI METODE ALGORITMA GENETIKA UNTUK MENENTUKAN PENJADWALAN PELAJARAN DI SMKN 1

Abstrak: Menentukan jadwal pelajaran merupakan suatu hal yang sangat penting dalam kegiatan mengajar. Dengan memiiki jadwal pelajaran yang tepat maka guru yang mengajar serta siswa dapat menjalani pelajaran dengan baik. Oleh karena itu guru maupun siswa memerlukan penjadwalan yang baik agar tidak terjadi bentrok antara pelajaran yang satu dengan pelajaran yang lainnya. Nah dalam kasus tersebut maka sangat diperluknnya sistem algoritma yang baik yaitu dengan menggunakan sistem algoritma genetika yang dapat membantu meringankan kesulitan penjadwalan tersebut. Metode ini lebih memudahkan kita melakukan penjadwalan dibandingkan metode konvensional. Dengan kita menerapkan metode ini untuk kedapannya maka akan sangat membantu kita untuk melakukan penjadwalan pelajaran disekolah SMKN 1 yang lebih efektif dan dan tapat.Kata kunci: Algoritma Genetika, Penjadwalan Pelajaran.Abstract: Determining a lesson schedule is very important in teaching activities. By having the right lesson schedule, teachers who teach and students can undergo lessons well. Therefore, teachers and students need fair scheduling to have no clash between one class and another. In this case, a suitable algorithm system is required, namely by using a genetic algorithm system that can help alleviate scheduling difficulty. This method makes scheduling more comfortable than conventional methods. Applying this method to its practicality will significantly help us schedule lessons at SMKN 1, which is more effective and accurate.Keywords: Genetic Algorithm, Lesson Scheduling.