Average Queuing Time Research Articles

Queuing Time Prediction Using WiFi Positioning Data in an Indoor Scenario.

Queuing is common in urban public places. Automatically monitoring and predicting queuing time can not only help individuals to reduce their wait time and alleviate anxiety but also help managers to allocate resources more efficiently and enhance their ability to address emergencies. This paper proposes a novel method to estimate and predict queuing time in indoor environments based on WiFi positioning data. First, we use a series of parameters to identify the trajectories that can be used as representatives of queuing time. Next, we divide the day into equal time slices and estimate individuals’ average queuing time during specific time slices. Finally, we build a nonstandard autoregressive (NAR) model trained using the previous day’s WiFi estimation results and actual queuing time to predict the queuing time in the upcoming time slice. A case study comparing two other time series analysis models shows that the NAR model has better precision. Random topological errors caused by the drift phenomenon of WiFi positioning technology (locations determined by a WiFi positioning system may drift accidently) and systematic topological errors caused by the positioning system are the main factors that affect the estimation precision. Therefore, we optimize the deployment strategy during the positioning system deployment phase and propose a drift ratio parameter pertaining to the trajectory screening phase to alleviate the impact of topological errors and improve estimates. The WiFi positioning data from an eight-day case study conducted at the T3-C entrance of Beijing Capital International Airport show that the mean absolute estimation error is 147 s, which is approximately 26.92% of the actual queuing time. For predictions using the NAR model, the proportion is approximately 27.49%. The theoretical predictions and the empirical case study indicate that the NAR model is an effective method to estimate and predict queuing time in indoor public areas.

Lining up for quick service—The business impact of express lines on fast-food restaurant operations

ABSTRACTThis article explores how waiting times at a fast-food restaurant can be modified. Toward the end of the article, a simulation model is proposed which examines queuing times in three different line scenarios based on the number of meal sets ordered. Data were collected using on-site observations of customer arrivals, waits, and service provision in a fast-food restaurant. Simulations were run via Simul8. Analysis was done using analysis of variance, as well as post-hoc tests. It was found that the waiting line configuration, arrival rate, and demand distribution all have significant impacts on average queuing time and weighted value.

Adaptive application-aware job scheduling optimization strategy in heterogeneous infrastructures

As the rapid advancement and diversity in the computing systems, it is demanding to take the most robust scheduling algorithm that guarantee an optimized performance to execute manifold applications on large-scale heterogeneous computing environments. This paper present an adaptive application-aware job scheduling optimization strategy for large-scale high throughput computing in heterogeneous infrastructures. The proposed scheduling optimization method is built on two main concepts. First, it provides application-aware job distribution weights through empirical data in large-scale heterogeneous infrastructures. Here we adopt the concept of weight, which represent the ratio of tasks that will be computed on each resource. The weights can vary in terms of application type and are optimized until the system get in steady status. Secondly, it offers an adaptive control phase that is invoked by the weight adjustment and resource scaling feature. The feedback data from monitoring module is forwarded to the control phase in order to adjust weights and over-provisioning ratio, and result in enhancing overall balance between performances and utilization of system. The experimental evaluation with the four realistic workload patterns demonstrates that, when compared to the core-based scheme which distributes tasks in proportions of each resource's number of cores, the use of our optimization method can achieve 62 % better average throughput, 43 % shorter average queueing time, and 38 % better average utilization of the entire resources in diverse infrastructure environments by harnessing our adaptive module.

Flexible Capacity Management with Future Information

We consider a maintenance service provider that overhauls aircraft engines in a central service facility. The service provider can choose between a low and a high capacity. However, the high capacity can only be used for a certain share of the time. We model the service facility as a queue with adaptable service rate and develop capacity control policies minimizing the time-average queue length. The reactive threshold-type policy only considers the current queue length and we show that the time-average queue length diverges as the load of the system increases. State-of-the-art aircraft engines are equipped with sensors permitting prediction of future maintenance needs. Thus, the solely forward-looking policy is based on predicted arrival times and maintenance requirements of engines arriving to the system in the future. We show that the time-average queue length converges to a finite value as the load increases. As the reactive policy outperforms the solely forward-looking policy for low arrival rates, we combine both to a proactive policy. While we assume that future information is available within an infinite lookahead window when stating the policy, we derive conditions under which a proactive capacity control policy can be employed and how the policy has to be modified if the lookahead window is finite.

Toward transcoding as a service: energy-efficient offloading policy for green mobile cloud

In this article we investigate energy-efficient offloading policy for transcoding as a service (TaaS) in a generic mobile cloud system. Computation on mobile devices can be offloaded to a mobile cloud system that consists of a dispatcher at the front end and a set of service engines at the back end. Particularly, a transcoding task can be executed on the mobile device (i.e. mobile execution) or offloaded and scheduled by the dispatcher to one of the service engines in the cloud (i.e. cloud execution). We aim to minimize the energy consumption of transcoding on the mobile device and service engines in the cloud while achieving low delay. For the mobile device, we formulate its offloading policy under delay deadline as a constrained optimization problem. We find an operational region on which execution mode, that is, mobile execution or cloud execution, is more energy efficient for the mobile device. For the cloud, we propose an online algorithm to dispatch transcoding tasks to service engines, with an objective to reduce energy consumption while achieving queue stability. By appropriately choosing the control variable, the proposed algorithm outperforms alternative algorithms, with lower time average energy consumption and time average queue length on the service engines. The proposed offloading policy can reduce energy consumption on both mobile devices and the cloud jointly, which provides guidelines for the design of green mobile cloud.

Queuing with future information

We study an admissions control problem, where a queue with service rate $1-p$ receives incoming jobs at rate $\lambda\in(1-p,1)$, and the decision maker is allowed to redirect away jobs up to a rate of $p$, with the objective of minimizing the time-average queue length. We show that the amount of information about the future has a significant impact on system performance, in the heavy-traffic regime. When the future is unknown, the optimal average queue length diverges at rate $\sim\log_{1/(1-p)}\frac{1}{1-\lambda}$, as $\lambda\to 1$. In sharp contrast, when all future arrival and service times are revealed beforehand, the optimal average queue length converges to a finite constant, $(1-p)/p$, as $\lambda\to1$. We further show that the finite limit of $(1-p)/p$ can be achieved using only a finite lookahead window starting from the current time frame, whose length scales as $\mathcal{O}(\log\frac{1}{1-\lambda})$, as $\lambda\to1$. This leads to the conjecture of an interesting duality between queuing delay and the amount of information about the future.

Toward Transcoding as a Service in a Multimedia Cloud: Energy-Efficient Job-Dispatching Algorithm

In this paper, we investigate the energy-efficient job-dispatching algorithm for transcoding as a service (TaaS) in a multimedia cloud. We aim to minimize the energy consumption of service engines in the cloud while achieving low delay for TaaS. We formulate the job-dispatching problem as a constrained optimization problem under the framework of Lyapunov optimization. Using the drift-plus-penalty function, we propose an online algorithm that dispatches the transcoding jobs to service engines, with an objective to Reduce Energy consumption while achieving the QUEue STability (REQUEST). We first characterize the fundamental tradeoff between energy consumption and queue delay for the REQUEST algorithm numerically and obtain its performance bound theoretically. Second, we study the robustness of the REQUEST algorithm, with numerical results indicating that the REQUEST algorithm is robust to the inaccuracy of estimating the transcoding time. Third, we compare the performance of the REQUEST algorithm with the other two algorithms, i.e., the Round Robin and Random Rate algorithms. By simulation and real trace data, we show that by appropriately choosing the control variable, the REQUEST algorithm outperforms the Round Robin and Random Rate algorithms, with smaller time average energy consumption and time average queue length. The proposed REQUEST algorithm can be applied in cloud-assisted multimedia transcoding service.

Queueing with future information

We study an admissions control problem, where a queue with service rate 1 - p receives incoming jobs at rate λ ε (1? p , 1), and the decision maker is allowed to redirect away jobs up to a rate of p , with the objective of minimizing the time-average queue length. We show that the amount of information about the future has a significant impact on system performance, in the heavy-traffic regime. When the future is unknown, the optimal average queue length diverges at rate ~ log 1/1- p 1/1-λ, as λ → 1. In sharp contrast, when all future arrival and service times are revealed beforehand, the optimal average queue length converges to a finite constant, (1 - p )/ p , as λ → 1. We further show that the finite limit of (1 - p )/ p can be achieved using only a finite lookahead window starting from the current time frame, whose length scales as O(log 1/1-λ), as λ → 1. This leads to the conjecture of an interesting duality between queuing delay and the amount of information about the future.

Editorial: Farewell 2013, Welcome 2014

In 2013, TSC expanded in a number of dimensions, including scope, editorial board, and contributing communities. First, TSC expanded its scope by introducing the new taxonomy of services computing (http://www.computer.org/ portal/web/tsc/taxonomy). This new taxonomy broadens the scope of TSC from the foundations of services computing to cover cloud computing services, big data technologies, the Internet of Things (IoT), social computing, and business management. This new taxonomy refl ects the continued advancements of services computing from both services science and services engineering perspectives. Second, TSC has expanded its editorial board to 43 associate editors in response to the broadening scope of TSC and the growing community of researchers and practitioners as authors and readers. More importantly, the TSC community continues to expand and attract both technical professionals and business management professionals. In recognition of its important role, TSC has been indexed by major indexing systems, such as Thomson Reuters (ISI), EI Compendex, and Science Citation Index Expanded (SCI-e). TSC welcomes 2014 with a number of new projects on the way. We are introducing survey papers as a featured section in TSC regular issues, in addition to special topic sections where guest editors are invited to solicit submissions on hot and emerging topics of special interests. We are planning to work with the steering committees of the IEEE International Conference on Web Services and the IEEE International Conference on Cloud Computing to create TSC special sessions. Authors of TSC accepted papers are offered an opt-in opportunity to present their work at one of these two conferences. The EiC is working closely with the TSC advisory board to explore new dimensions to attract higher readership and authorship from the business and management community. Finally, The EiC will continue to work with the TSC editorial board through multiple threads to maintain the average review cycle of three months and the average queuing time of less than one year for accepted papers to appear. One new thread of such effort is to continue TSC's participation in the newly introduced Reviewer Appreciation Program by the IEEE Computer Society.

An integrated release and dispatch policy for semiconductor wafer fabrication

This paper presents an integrated release and dispatch policy for semiconductor wafer fabrication. In this policy, the concept of virtual production lines (VPLs) is first introduced to allocate production time for each process step, and the average queuing time information is derived from the target WIP distribution which is obtained by simulations. Then, a pull VPLs-based effective-workload control (EWL-n-Ctrl) policy is proposed to control the release process, and a queuing time ratio (QTR) rule is put forward to control the dispatch process. The behaviour of EWL-n-Ctrl is analysed using the Mini-Fab model. Results of the numerical experiments show that the integrated release and dispatch policy (EWL-n-Ctrl+QTR) outperforms common rule-based scheduling policies in terms of the average cycle time, the standard deviation of the cycle time and the work in process under a prescribed throughput level. A simulation study into the effect of capacity utilisation levels and product mix ratios is also carried out.

A New Method to Improve Round Robin Scheduling Algorithm with Quantum Time Based on Harmonic-Arithmetic Mean (HARM)

One of the most important concepts in multi programming Operating Systems is scheduling. It helps in choosing the processes for execution. Round robin method is one of the most important algorithms in scheduling. It is the most popular algorithm due to its fairness and starvation free nature towards the processes, which is achieved by using proper quantum time. The main challenge in this algorithm is selection of quantum time. This parameter affects on average Waiting Time and average Turnaround Time in execution queue. As the quantum time is static, it causes less context switching in case of high quantum time and high context switching in case of less quantum time. Increasing context switch leads to high average waiting time, high average turnaround time which is an overhead and degrades the system performance. With respect to these points, the algorithms should calculate proper value for the quantum time. Two main classes of algorithms that are proposed to calculate the quantum time include static and dynamic methods. In static methods quantum time is fixed during the scheduling. Dynamic algorithms are one of these methods that change the value of quantum time in each cycle. For example in one method the value of quantum time in each cycle is equal to the median of burst times of processes in ready queue and for another method this value is equal to arithmetic mean of burst times of ready processes. In this paper we proposed a new method to obtaining quantum time in each cycle based on arithmetic- harmonic mean (HARM). Harmonic mean is calculated by dividing the number of observations by the reciprocal of each number in the series. With examples we show that in some cases it can provides better scheduling criteria and improves the average Turnaround Time and average Waiting Time.

Queueing System with On-Demand Number of Servers

We consider a queuing system where the number of active servers changes depending on the length of the queue. As a practical example of such system, we consider the security check queue at the airports. The number of active servers increases when the queue grows by k customers and decreases accordingly. That allows to save server resources while maintaining acceptable performance (average queuing time and its variation) for customers. We obtain a closed-form solution for the serving time, queue length and average number of servers. To validate the model we have selected the data of Dallas Fort international airport, the 8th largest in the world. Our simulation model shows a close match with analytic results. Cost savings in the number of open servers are achievable while providing acceptable waiting time for the customers.

Onshore ballast water treatment: A viable option for major ports

Ballast water treatment consists of the elimination of exotic species. Currently, the development of alternative methods for this process is directed toward treatment onboard ships. However, we present onshore treatment as a viable alternative for ballast water treatment. We investigated onshore treatment in two iron ore ports with movement capacities of 25 and 90million tons annually (Mta) that receive 7.5 and 25million cubic meters annually (Mm3) of ballast water, respectively. Discrete event simulation was used as the method of analysis, considering the processes of arrival, berthing, ship loading and capture and treatment of ballast water. We analyzed data from 71 ships operating in these ports to validate our simulation model. We were able to demonstrate that onshore treatment does not impact the cargo capacity, occupation rate or average queuing time of ships at these ports. We concluded that implementation of onshore ballast water treatment may be practicable in ports that receive high volumes of ballast water.

Investigating the effect of collective organizational commitment on unit‐level performance and absence

The relationship between individual‐level organizational commitment and employee's performance has been the subject of considerable empirical investigation over several decades. However, few studies have examined the effect of collective organizational commitment on unit or organizational‐level performance, even though available theory suggests commitment's effects may be stronger at the unit level. We present results of a study examining the effects of unit‐level organizational commitment on the speed and quality of performance, and employee absence, drawing on a sample of 893 service sector employees (representing a 90% response rate) from 39 office units in a UK public sector organization. Results show that unit‐level organizational commitment was statistically significantly associated with both unit‐level performance quality (i.e., reduced customer complaints per service transaction) and performance speed (i.e., customer average queuing time). Collective organizational commitment was not significantly associated with a unit‐level measure of the average time customers spent at service counters with the tellers, nor was it associated with unit‐level employee absence.Practitioner Points Bottom‐line impact of organizational commitment: unit‐level organizational commitment had substantial effects on unit‐level performance. Managers must recognize that committed employees have a value beyond their immediate and individual performance as such individuals contribute to unit performance. Unit‐level organizational commitment implies different managerial interventions. Our research indicates the value to organizations of analysing survey data at the unit‐level: it is easier to ensure participant anonymity and confidentiality of responses and it provides information beyond individual level data.

Summer internships in sigma-sw

This past summer I spent eight weeks working with Cardiff University School of Mathematics undertaking a research project measuring the efficiency of Mathematics Support Centres. This was achieved by using computer simulation to model a Mathematics Support Centre and observe various performance parameters such as: ‘time spent in the centre’, ‘average queuing time’ and ‘the number of students who successfully exit the system having received a desired amount of support’. A subsidiary aim of the project was to produce a user-friendly online resource, which could be 1. Ollie Bond Mathematics University of Exeter ob235@ex.ac.uk

Practical insights on pitfalls in estimating queue performance

Significant estimation errors, especially in high utilisation or low sample size situations, can occur when using sample data in standard queueing formulas to estimate queueing measures like time in queue. Perhaps surprisingly, these issues are rarely addressed, and especially in materials developed for students and practitioners. We establish error bounds in the case of exponential arrivals and Poisson service times for the single-server system when estimating average time in queue and also provide practical guidance for practitioners.

Performance Evaluation and Dimensioning of GIX/M/c/N Systems through Kernel Estimation

We extend the analysis of queueing systems for real‐life situations, where the arrival pattern of customers is unknown. In real systems, we must understand how the choice of a method of estimation influences the configuration of the system. Using kernel smoothing, we evaluate algorithms to estimate performance measures of a GIX/M/c/N system, including the invariant probability distribution of the number of customers in the system, the blocking probability, the average queue size, and the average client queue time. We successfully apply the method to the arrivals to a call center to plan and improve the performance of these important queueing systems.

The average response time in a heavy-traffic srpt queue

Shortest Remaining Processing Time first (SRPT) has long been known to optimize the queue length distribution and the mean response time (a.k.a. flow time, sojourn time). As such, it has been the focus of a wide body of analysis. However, results about the heavy-traffic behavior of SRPT have only recently started to emerge. In this work, we characterize the growth rate of the mean response time under SRPT in the M/GI/1 system under general job size distributions. Our results illustrate the relationship between the job size tail and the heavy traffic growth rate of mean response time. Further, we show that the heavy traffic growth rate can be used to provide an accurate approximation for mean response time outside of heavy traffic.

A Dispatching Rule Study of a LED Sorting System

The scheduling problem of LED chip sorting workstation is a multiprocessor open shop scheduling problem. In practice, this problem is usually solved with finding a dispatching rule to achieve certain performance indexes. Experimental design was applied to find a proper rule, and dispatching rule, wafer quantity, wafer mix type, and bottleneck location were chosen as independent variables. Maximum make-span and average queuing time were selected as dependent variables. Five dispatching rules were considered in the dispatching rule factor. Experimental results showed that all independent variables are significant on all dependent variables. While make-span as the performance index, three dispatching rules common in applying LPT in bottleneck machines obtained better results. A dispatching rule with applying SPT in non-bottleneck machines has least average queuing time among those three, and this rule is recommended for practical application.

Performance analysis of immigration operation by discrete event modelling and simulation

Discrete event modelling and simulation were used to analyse the performance of immigration operation in Botswana. The relationships between length of queues of immigrants, queuing time, service time and engagem-ent of duty officer were investigated. Data collected by direct observation and clock-timing of processing immi-grant request at an inland office on a normal working day was used to determine whether to increase or reduce the number of serving officers to balance the operations. The findings indicated that the system of operation was balanced or fairly matched by 75 % utilization of the officer, average immigrant queuing time of 4.2 min, and required no changes. The pilot study could be replicated at other boarder posts and airports prior to the expected influx of tourists during the 2010 World Cup Football Competition in South Africa.