Time-varying Demand Patterns Research Articles

Power Quality Impacts of Grid-Tied PV Inverters on Low Voltage Distribution Networks a Smart OpenDSS Model to Find Power Quality Threshold Limits

As a renewable initiative, risen attention can be seen in solar photovoltaic (PV) installations in the world. Since tropical countries and even many other countries during summer experience higher irradiance levels, there is great potential for PV generation in solar panels. However, power quality can be impacted when the number of PV modules is increased in a network due to the emergence of harmonics, unbalanced voltages, voltage flicker, neutral voltage variations, etc. Therefore, it is essential to empirically analyse power quality parameters, i.e., total harmonic distortion (THD), voltage unbalance, etc., and find threshold margins of PV capacities in a network. Based on a case study conducted in Negombo, Sri Lanka, this work determines power quality impacts with reference to the number of PV interconnections in a distributed network. For the task, a low-voltage distribution network was chosen which was fed by a 250kVA, 11/0.4kV transformer with domestic loads and grid-tied PV inverters. A simulation model was developed in Open Distribution System Simulator (OpenDSS) with time-varying load patterns and PV generation. Consequently, a smart efficient method was introduced to model domestic loads with unique time-varying demand patterns. A determination criterion was established to derive snapshot load flow instants using time-varying load flow results to analyse voltage profiles along distribution feeders. Then, the model was enhanced to quantify the power quality parameters such as individual harmonic content, THD of voltage as well as current and neutral voltage variation. The results reveal that node voltage has improved with PV interception without violating the upper limit. The THD of voltage and current have slightly increased with the addition of PV inverters.

An agent-based algorithm for dynamic routing in service networks

Service networks consisting of components and links are regarded as significant infrastructures to provide all kinds of services for users. They are supposed to select the best route for service requests in case of time-varying demand and different service patterns. In addition, it is necessary to choose appropriate routing metrics so as to satisfy user requirements for a variety of services. Multiple quality of service (QoS) requirements must be considered in the dynamic environment as a result of high demand for service networks. To that end, we propose an agent-based algorithm to address the routing problem in service networks. We construct a multi-layer network model to figure out component behaviors and complicated relationship between components under uncertainty. In order to reflect various service requirements, QoS metrics are defined from the perspectives of component and link. We also put forward an improved deep Q-learning method to achieve global convergence and enhance the efficiency of the routing algorithm. The numerical results on a case study illustrate the proposed algorithm finds high-quality solutions at acceptable costs, which routes service requests properly in the dynamic network environment. The proposed algorithm achieves outstanding performance compared with state-of-the-art routing algorithms in terms of delay and service factor.

Self-Learning Threshold-Based Load Balancing

We consider a large-scale service system where incoming tasks have to be instantaneously dispatched to one out of many parallel server pools. The user-perceived performance degrades with the number of concurrent tasks and the dispatcher aims at maximizing the overall quality of service by balancing the load through a simple threshold policy. We demonstrate that such a policy is optimal on the fluid and diffusion scales, while only involving a small communication overhead, which is crucial for large-scale deployments. In order to set the threshold optimally, it is important, however, to learn the load of the system, which may be unknown. For that purpose, we design a control rule for tuning the threshold in an online manner. We derive conditions that guarantee that this adaptive threshold settles at the optimal value, along with estimates for the time until this happens. In addition, we provide numerical experiments that support the theoretical results and further indicate that our policy copes effectively with time-varying demand patterns. Summary of Contribution: Data centers and cloud computing platforms are the digital factories of the world, and managing resources and workloads in these systems involves operations research challenges of an unprecedented scale. Due to the massive size, complex dynamics, and wide range of time scales, the design and implementation of optimal resource-allocation strategies is prohibitively demanding from a computation and communication perspective. These resource-allocation strategies are essential for certain interactive applications, for which the available computing resources need to be distributed optimally among users in order to provide the best overall experienced performance. This is the subject of the present article, which considers the problem of distributing tasks among the various server pools of a large-scale service system, with the objective of optimizing the overall quality of service provided to users. A solution to this load-balancing problem cannot rely on maintaining complete state information at the gateway of the system, since this is computationally unfeasible, due to the magnitude and complexity of modern data centers and cloud computing platforms. Therefore, we examine a computationally light load-balancing algorithm that is yet asymptotically optimal in a regime where the size of the system approaches infinity. The analysis is based on a Markovian stochastic model, which is studied through fluid and diffusion limits in the aforementioned large-scale regime. The article analyzes the load-balancing algorithm theoretically and provides numerical experiments that support and extend the theoretical results.

An enhanced simulation-based iterated local search metaheuristic for gravity fed water distribution network design optimization

The gravity fed water distribution network design (WDND) optimization problem consists in determining the pipe diameters of a water network such that hydraulic constraints are satisfied and the total cost is minimized. Traditionally, such design decisions are made on the basis of expert experience. When networks increase in size, however, rules of thumb will rarely lead to near optimal decisions. Over the past thirty years, a large number of techniques have been developed to tackle the problem of optimally designing a water distribution network. In this paper, we tackle the NP-hard WDND optimization problem in a multi-period setting where time varying demand patterns occur. We propose a new enhanced simulation-based iterated local search (ILS) metaheuristic which further explores the structure of the problem in an attempt to obtain high quality solutions. More specifically, four novelties are proposed: (a) a local search strategy to smartly dimension pipes in the shortest paths between the reservoirs and the nodes with highest demands; (b) a technique to speed up convergence based on an aggressive pipe diameter reduction scheme; (c) a novel concentrated perturbation mechanism to allow escaping from very restrained local optima solutions; and (d) a pool of solutions to achieve a good compromise between intensification and diversification. Computational experiments show that our approach is able to improve over a state-of-the-art metaheuristic for most of the performed tests. Furthermore, it converges much faster to low cost solutions and demonstrates a more robust performance in that it obtains smaller deviations from the best known solutions.

WiLiTV: Reducing Live Satellite TV Costs Using Wireless Relays

The bandwidth required for TV content distribution is rapidly increasing due to the evolution of high definition TV (HDTV) and ultra HDTV. Service providers are constantly trying to differentiate themselves by innovating new ways of distributing content more efficiently with lower cost and higher penetration. We propose a cost-efficient wireless architecture [wireless live TV (WiLiTV)], consisting of a mix of wireless access technologies [satellite, Wi-Fi, and LTE/5G millimeter wave (mmWave) overlay links], for delivering live TV services. In the proposed architecture, live TV content is injected into the network at selected locations, consisting of some homes and/or cellular base stations, using satellite antennas. The content is then further distributed to other homes using a house-to-house Wi-Fi network or an LTE/5G mmWave overlay. We construct an optimal content distribution network with the minimum number of satellite injection points, while preserving the highest quality of experience, for different neighborhood densities. We evaluate the framework using time-varying demand patterns and a diverse set of home location data provided from an operational content distribution network. Our study demonstrates that this architecture reduces the overall cost by 60% compared with the traditional architecture. We have also shown that the WiLiTV is robust in its support for several TV formats.

Optimal Service Elasticity in Large-Scale Distributed Systems

A fundamental challenge in large-scale cloud networks and data centers is to achieve highly efficient server utilization and limit energy consumption, while providing excellent user-perceived performance in the presence of uncertain and time-varying demand patterns. Auto-scaling provides a popular paradigm for automatically adjusting service capacity in response to demand while meeting performance targets, and queue-driven auto-scaling techniques have been widely investigated in the literature. In typical data center architectures and cloud environments however, no centralized queue is maintained, and load balancing algorithms immediately distribute incoming tasks among parallel queues. In these distributed settings with vast numbers of servers, centralized queue-driven auto-scaling techniques involve a substantial communication overhead and major implementation burden, or may not even be viable at all.Motivated by the above issues, we propose a joint auto-scaling and load balancing scheme which does not require any global queue length information or explicit knowledge of system parameters, and yet provides provably near-optimal service elasticity. We establish the fluid-level dynamics for the proposed scheme in a regime where the total traffic volume and nominal service capacity grow large in proportion. The fluid-limit results show that the proposed scheme achieves asymptotic optimality in terms of user-perceived delay performance as well as energy consumption. Specifically, we prove that both the waiting time of tasks and the relative energy portion consumed by idle servers vanish in the limit. At the same time, the proposed scheme operates in a distributed fashion and involves only constant communication overhead per task, thus ensuring scalability in massive data center operations. Extensive simulation experiments corroborate the fluid-limit results, and demonstrate that the proposed scheme can match the user performance and energy consumption of state-of-the-art approaches that do take full advantage of a centralized queue.

Optimal Service Elasticity in Large-Scale Distributed Systems

A fundamental challenge in large-scale cloud networks and data centers is to achieve highly efficient server utilization and limit energy consumption, while providing excellent user-perceived performance in the presence of uncertain and time-varying demand patterns. Auto-scaling provides a popular paradigm for automatically adjusting service capacity in response to demand while meeting performance targets, and queue-driven auto-scaling techniques have been widely investigated in the literature. In typical data center architectures and cloud environments however, no centralized queue is maintained, and load balancing algorithms immediately distribute incoming tasks among parallel queues. In these distributed settings with vast numbers of servers, centralized queue-driven auto-scaling techniques involve a substantial communication overhead and major implementation burden, or may not even be viable at all. Motivated by the above issues, we propose a joint auto-scaling and load balancing scheme which does not require any global queue length information or explicit knowledge of system parameters, and yet provides provably near-optimal service elasticity. We establish the fluid-level dynamics for the proposed scheme in a regime where the total traffic volume and nominal service capacity grow large in proportion. The fluid-limit results show that the proposed scheme achieves asymptotic optimality in terms of user-perceived delay performance as well as energy consumption. Specifically, we prove that both the waiting time of tasks and the relative energy portion consumed by idle servers vanish in the limit. At the same time, the proposed scheme operates in a distributed fashion and involves only constant communication overhead per task, thus ensuring scalability in massive data center operations. Extensive simulation experiments corroborate the fluid-limit results, and demonstrate that the proposed scheme can match the user performance and energy consumption of state-of-the-art approaches that do take full advantage of a centralized queue.

An Iterated Local Search Algorithm for Multi-Period Water Distribution Network Design Optimization

Water distribution networks consist of different components, such as reservoirs and pipes, and exist to provide users (households, agriculture, industry) with high-quality water at adequate pressure and flow. Water distribution network design optimization aims to find optimal diameters for every pipe, chosen from a limited set of commercially available diameters. This combinatorial optimization problem has received a lot of attention over the past forty years. In this paper, the well-studied single-period problem is extended to a multi-period setting in which time varying demand patterns occur. Moreover, an additional constraint—which sets a maximum water velocity—is imposed. A metaheuristic technique called iterated local search is applied to tackle this challenging optimization problem. A full-factorial experiment is conducted to validate the added value of the algorithm components and to configure optimal parameter settings. The algorithm is tested on a broad range of 150 different (freely available) test networks.

Economic production quantity model with trade credit financing and price-discount offer for non-decreasing time varying demand pattern

As a policy when some suppliers offer trade credit periods and price discounts to the retailers in order to increase the demand of their product, the retailers have to face different types of offer of discounts and credits for which they have to take a decision which is the best offer for them for profit. One thing is important to the retailers that they try to buy good quality items at a reasonable price and if possible, they try to invest returns obtained by selling those items in such a manner that their business is not hampered. In this paper, we intend to develop an economic production quantity (EPQ)-based model with non-decreasing time varying (quadratic) demand pattern in order to investigate the inventory system as a profit maximisation problem when delay in payment and price discount are permitted by the suppliers to the retailers. Mathematical theorems are developed analytically to determine optimal replenishment policies and a lot of managerial phenomena are obtained through numerical examples.

Optimality analysis of energy-performance trade-off for server farm management

A central question in designing server farms today is how to efficiently provision the number of servers to extract the best performance under unpredictable demand patterns while not wasting energy. While one would like to turn servers off when they become idle to save energy, the large setup cost (both, in terms of setup time and energy penalty) needed to switch the server back on can adversely affect performance. The problem is made more complex by the fact that today’s servers provide multiple sleep or standby states which trade off the setup cost with the power consumed while the server is ‘sleeping’. With so many controls, finding the optimal server farm management policy is an almost intractable problem — How many servers should be on at any given time, how many should be off, and how many should be in some sleep state? In this paper, we employ the popular metric of Energy-Response time Product (ERP) to capture the energy-performance trade-off, and present the first theoretical results on the optimality of server farm management policies. For a stationary demand pattern, we prove that there exists a very small, natural class of policies that always contains the optimal policy for a single server, and conjecture it to contain a near-optimal policy for multi-server systems. For time-varying demand patterns, we propose a simple, traffic-oblivious policy and provide analytical and empirical evidence for its near-optimality.

Calibration of Microscopic Traffic Simulation Models

A mathematical framework and a solution approach are presented for the simultaneous calibration of the demand and supply parameters and inputs to microscopic traffic simulation models as well as a large-scale application emphasizing practical issues. Microscopic traffic simulation models provide detailed estimates of evolving network conditions by modeling time-varying demand patterns and individual drivers' detailed behavioral decisions. Such models are composed of elements that simulate different demand and supply processes and their complex interactions. Several model inputs (such as origin-destination flows) and parameters (car-following and lane-changing coefficients) must be specified before these simulation tools can be applied, and their values must be determined so that the simulation output accurately replicates the reality reflected in traffic measurements. A methodology is presented here for simultaneously estimating all microscopic simulation model parameters by using general traffic measurements. A large-scale case study for the calibration of the MITSimLab microscopic traffic simulation model by using the network of Lower Westchester County, New York, is employed to demonstrate the feasibility, application, and benefits of the proposed methodology.

Performance Metrics for Advanced Access

Advanced access is an outpatient scheduling technique that aims to provide sameday appointment access. It is designed to reduce the time patients must wait for a scheduled appointment and to improve continuity of care by matching daily appointment supply and demand. Factors that make it difficult to sustain initial success in achieving supply-demand balance include different practice styles of doctors, differences in panel compositions and patient preferences, and time-varying demand patterns. This article proposes several performance measures that can help clinic directors monitor and evaluate their advanced access implementation. We also discuss strategies for sustaining advanced access in the long run.

Economic order quantity models for ameliorating/deteriorating items under inflation and time discounting

The items that incur a gradual loss in quality or quantity over time while in inventory are usually called deteriorating items. In reality, there are some items whose value or utility or quantity increase with time and those items can be termed as ameliorating items. In this paper, an effort has been made to incorporate these two opposite physical characteristics of stored items into inventory model. We develop models for ameliorating/deteriorating items with time-varying demand pattern over a finite planning horizon, taking into account the effects of inflation and time value of money. Optimal solutions of the proposed models are derived and the effects of amelioration/deterioration on the inventory replenishment policies are studied with the help of numerical examples.

A note on an order-level inventory model for a deteriorating item with time-dependent quadratic demand

An order-level inventory problem is discussed with the demand rate being represented by a continuous, quadratic function of time. It is assumed that a constant fraction of the on-hand inventory deteriorates per unit of time. The solution of the model is discussed both for infinite and finite time-horizon. A numerical example is taken up to illustrate the solution procedure and sensitivity analysis is also carried out. The rationale for the time-dependent quadratic demand is discussed. Scope and purpose The purpose of the present paper is to give a new dimension to the inventory literature on time-varying demand patterns. Researchers have extensively discussed various types of inventory models with linear trend (positive or negative) in demand. The main limitation in linear time-varying demand rate is that it implies a uniform change in the demand rate per unit time. This rarely happens in the case of any commodity in the market. In recent years, some models have been developed with a demand rate that changes exponentially with time. Demands for spare parts of new aeroplanes, computer chips of advanced computer machines, etc. increase very rapidly while the demands for spares of the obsolete aeroplanes, computers etc. decrease very rapidly with time. Some modellers suggest that this type of rapid change in demand can be represented by an exponential function of time. The present authors feel that an exponential rate of change in demand is extraordinarily high and the demand fluctuation of any commodity in the real market cannot be so high. A realistic approach is to think of accelerated growth (or decline) in the demand rate in the situations cited above and it can be best represented by a quadratic function of time. Thus, this paper has the scope of direct application in the very practical situations noted above.

Using MLP networks to design a production scheduling system

This paper investigates the application of artificial neural networks to the problem of job shop scheduling with a scope of a deterministic time-varying demand pattern over a fixed planning horizon. The purpose of the research is to design and develop a job shop scheduling system (a scheduling software) that can generate effective job shop schedules using the multi-layered perceptron (MLP) networks. The contributions of this study include designing, developing, and implementing a production activity scheduling system using the MLP networks; developing a method for organizing sample data using a denotation bit to indicate processing sequence and processing time of a job simultaneously; using the back-propagation training process to control local minimal solutions; and developing a heuristics to improve and revise the initial production schedule. The proposed production activity schedule system is tested in a real production environment and illustrated in the paper with a sample case.

A note on a lot sizing heuristic for deteriorating items with time-varying demands and shortages

In the present article, an existing heuristic procedure is applied to an inventory of deteriorating items with (i) linearly and (ii) exponentially time-varying demands and shortages over a fixed planning horizon. In a similar problem, Hariga and Al-Alyan [Hariga M, Al-Alyan A. A lot sizing heuristic for deteriorating items with shortages in growing and declining markets. Computers & Operations Research 1997;24:1075–83] allowed shortages in all cycles except in the last one ( policy-B). Here shortages are considered in each cycle and are completely backlogged ( policy-A). The successive scheduling periods over an infinite time horizon are adjusted for a finite planning horizon following different adjustment policies (APs) that are proposed in an algorithmic form. Numerical examples are taken from Hariga and Al-Alyan [20] to show that the replenishment policy-A results in a lower average system cost than that of policy-B in both types of demand patterns. It is also found that the approximate results under the policy-A are closely associated with those of the optimal ones. Scope and purpose The purpose of the article is to implement an existing heuristic procedure to an inventory system of deteriorating items allowing shortages in all cycles over a finite planning horizon under linearly and exponentially time-varying demand patterns. Allowing shortages in all cycles is a significant departure from the traditional practice of allowing shortages in all cycles except the last one. With this departure, we reconsider the problem of Hariga and Al-Alyan [20] published recently in the Journal of Computers & Operations Research and find a reduction in the system cost upto 15%. The scope of applications of the heuristic is widely open for newly developed high technology products (e.g., microcomputer chips, sophisticated instruments, etc.) which have growing market demands or in the case of technologically obsolete products (e.g., spare parts of obsolete machines, etc.) having a gradually declining market demands.

An Algorithm for the Single-item Capacitated Lot-Sizing Problem with Concave Production and Holding Costs

We consider a multi-period single-item production scheduling problem with a deterministic, time-varying demand pattern and concave cost functions. Optimal production lot sizes, so as to minimize the total costs of production, set-up, and inventory, are determined subject to dynamic production capacity and no backlogs. The proposed algorithm was tested extensively by solving several randomly generated problems with varying degrees of complexity. The proposed algorithm appears to perform quite reasonably for practical applications.

On the discrete time-proportional demand inventory problem

This paper discusses a lot-size model with a very specific time-varying demand pattern, namely a time proportional demand which occurs only at discrete and equal intervals of time throughout a given planning horizon. The complexity and importance of the problem were recognized by Sasieni, Yaspan, and Friedman (1959), but remained basically ignored until Jaiswal and Shah (1979) provided some heuristics under the simplifying assumptions of equal intervals between consecutive orders and, in one case, equal lot sizes. More recently, Arcelus and Srinivasan have developed an optimal solution algorithm (1985a) as well as some heuristics (1985b) for the more general problem, where both the intervals between orders and the lot sizes are allowed to vary simultaneously. Three basic aspects of the discrete demand case will be considered in this paper, namely, (i) an examination of the characteristics that differentiate the discrete from the continuous demand cases; (ii) a discussion of the ways in which these characteristics can be exploited to find solution algorithms and heuristics for the discrete case; and (iii) a report on computational experience with these models. The comparison of the characteristics of the discrete and the continuous cases is useful for the potential contribution of the latter and better known problem to the solution of the former. Examples of work devoted to the continuous case during the last

An Algorithm for the Dynamic Lot-Size Problem with Time-Varying Production Capacity Constraints

We consider the situation of a single item having a deterministic, time-varying demand pattern. Production lot sizes, so as to minimize the total of setup and carrying costs, are to be determined subject to production capacity restrictions that can vary with time. Two important theorems concerning properties of the optimal solution are exploited in developing a tree search algorithm. In addition, other properties are utilized to reduce the amount of search necessary. Results of extensive tests are presented and discussed.