Length Of Time Horizon Research Articles

A Predictive Energy Management Strategy for Heavy Hybrid Electric Vehicles Based on Adaptive Network-Based Fuzzy Inference System-Optimized Time Horizon

Energy management strategies play a crucial role in enhancing the fuel efficiency of hybrid electric vehicles (HEVs) and mitigating greenhouse gas emissions. For the current commonly used time horizon optimization methods that only target the trend curve of the optimal battery state of charge (SOC) trajectory obtained offline, which are only suitable for buses with known future driving conditions, this paper proposed an energy management strategy based on an adaptive network-based fuzzy inference system (ANFIS) that optimizes the time horizon length and enhances adaptability to driving conditions by integrating historical vehicle velocity, accelerations, and battery SOC trajectory. First, the vehicle velocity prediction model based on the radial basis function (RBF) neural network is used to predict the future velocity sequence. After that, ANFIS was used to optimize and update the length of the forecast time horizon based on the historical vehicle velocity sequence. Finally, compared with the fixed time horizon energy management strategy, which is based on model predictive control (MPC), the average calculation time of the energy management strategy is reduced by about 23.5%, and the fuel consumption per 100 km is reduced by about 6.12%.

An Intelligent Approach to Short-Term Wind Power Prediction Using Deep Neural Networks

Abstract In this paper, an intelligent approach to the Short-Term Wind Power Prediction (STWPP) problem is considered, with the use of various types of Deep Neural Networks (DNNs). The impact of the prediction time horizon length on accuracy, and the influence of temperature on prediction effectiveness have been analyzed. Three types of DNNs have been implemented and tested, including: CNN (Convolutional Neural Networks), GRU (Gated Recurrent Unit), and H-MLP (Hierarchical Multilayer Perceptron). The DNN architectures are part of the Deep Learning Prediction (DLP) framework that is applied in the Deep Learning Power Prediction System (DLPPS). The system is trained based on data that comes from a real wind farm. This is significant because the prediction results strongly depend on weather conditions in specific locations. The results obtained from the proposed system, for the real data, are presented and compared. The best result has been achieved for the GRU network. The key advantage of the system is a high effectiveness prediction using a minimal subset of parameters. The prediction of wind power in wind farms is very important as wind power capacity has shown a rapid increase, and has become a promising source of renewable energies.

Dynamic horizon selection methodology for model predictive control in buildings

The interest in model predictive control (MPC) for buildings has grown in recent years due to the widespread implementation of dynamic electricity tariffs, energy flexibility and distributed energy resources. The MPC applied on buildings is a computational-based methodology used to optimize the performance of heating, ventilation and air conditioning systems (HVAC) by predicting the energy behavior and minimizing a specific cost function in a determined forecasting horizon. The forecasting horizon is one of the critical parameters in MPC design applied in buildings; it should be long enough to activate the buildings’ flexibility potential, but the computational resources grow exponentially with the horizon increase, which could difficult the real-time operation. Furthermore, long periods of non-occupancy, holidays or abrupt comfort-bound changes can significantly affect the optimal forecasting time horizon length. Unfortunately, very few studies have focused on ascertaining this key optimization process aspect. The contribution of this research paper is to demonstrate, through an innovative methodology, that the optimal horizon length can be dynamically updated according to the effects of building inertia. This methodology is validated by assessing the reduction of the economic costs of a space heating system based on a synthetic representation of an experimental building placed in Germany.

Aggregation of Demand-Side Flexibilities: A Comparative Study of Approximation Algorithms

Traditional power grids are mainly based on centralized power generation and subsequent distribution. The increasing penetration of distributed renewable energy sources and the growing number of electrical loads is creating difficulties in balancing supply and demand and threatens the secure and efficient operation of power grids. At the same time, households hold an increasing amount of flexibility, which can be exploited by demand-side management to decrease customer cost and support grid operation. Compared to the collection of individual flexibilities, aggregation reduces optimization complexity, protects households’ privacy, and lowers the communication effort. In mathematical terms, each flexibility is modeled by a set of power profiles, and the aggregated flexibility is modeled by the Minkowski sum of individual flexibilities. As the exact Minkowski sum calculation is generally computationally prohibitive, various approximations can be found in the literature. The main contribution of this paper is a comparative evaluation of several approximation algorithms in terms of novel quality criteria, computational complexity, and communication effort using realistic data. Furthermore, we investigate the dependence of selected comparison criteria on the time horizon length and on the number of households. Our results indicate that none of the algorithms perform satisfactorily in all categories. Hence, we provide guidelines on the application-dependent algorithm choice. Moreover, we demonstrate a major drawback of some inner approximations, namely that they may lead to situations in which not using the flexibility is impossible, which may be suboptimal in certain situations.

Stagewise Newton Method for Dynamic Game Control With Imperfect State Observation

In this letter, we study dynamic game optimal control with imperfect state observations and introduce an iterative method to find a local Nash equilibrium. The algorithm consists of an iterative procedure combining a backward recursion similar to minimax differential dynamic programming and a forward recursion resembling a risksensitive Kalman smoother. A coupling equation renders the resulting control dependent on the estimation. In the end, the algorithm is equivalent to a Newton step but has linear complexity in the time horizon length. Furthermore, a merit function and a line search procedure are introduced to guarantee convergence of the iterative scheme. The resulting controller reasons about uncertainty by planning for the worst case disturbances. Lastly, the low computational cost of the proposed algorithm makes it a promising method to do output-feedback model predictive control on complex systems at high frequency. Numerical simulations on realistic robotic problems illustrate the risk-sensitive behavior of the resulting controller.

Coping with uncertainty: Exploring the foresight actions’ role in supporting growth-orientation among Finnish dairy farmers

The two objectives of this study are to investigate how Finnish dairy farmers cope with uncertainties of their external business environment and how they apply foresight methods in their strategic management, e.g., in planning for expansion. The research data consist of 135 survey responses from larger than average dairy farmers across Finland. To reveal the role of future thinking and long-term management in farmers’ goal setting, we first carried out an explorative factor analysis (EFA) for the management related claims. The identified latent factors were managerial, planning, and cautious approach. Significant differences in managerial approach were found with respect to turnover and age of farmers. Planning approach correlated with farmer’s time horizon. In the binary logistic regression for predicting the growth orientation farmer’s planning approach, age and length of time horizon were the only significant predictors. Many farmers are particularly cautious when it comes to planning the future of milk production. Strengthening farmers’ foresight skills with training may help farmers to cope with volatile business environments.

Transitions among states behind interactive agent model

In this paper, we introduce a simple interactive agent mechanism, where the distribution of returns generated from the mechanism match stylized facts in financial markets. We introduce one more key factor, the length of time horizon on performance evaluations between strategies, which also has a significant influence on price fluctuations. To investigate the transitions among states, we introduce a Markov transition matrix, Perron‐Frobenius transition matrix, and Inertia. Our simulation results show the stickiness of states switching from one to another, and the longer length of time horizon on performance evaluations would generate more complex dynamic price fluctuations. We link our simple heterogeneous agent mechanism with Markov trajectory entropy and provide a total score and probability density functions of representations under two states as applications for the mechanism.

What Drives Robo-Advice?

The promise of robo-advisory firms is to provide low cost access to diversified portfolios built in accordance with the academic literature on normative portfolio choice. We investigate the latter claim. How much normative advice does robo-advice contain? For this purpose we web-scrap portfolio recommendations for 151200 investor types (input combinations from an online questionnaire) for one of the largest US robo-advisors. Our results show that the type of investment goal and the length of time horizon are dominating inputs with significant influence on recommended equity allocations. Normative advice in the form of Merton type hedging demands plays no role at all.

A Re-Solving Heuristic with Uniformly Bounded Loss for Network Revenue Management

We consider a canonical quantity-based network revenue management problem where a firm accepts or rejects incoming customer requests irrevocably in order to maximize expected revenue given limited resources. Because of the curse of dimensionality, the exact solution to this problem by dynamic programming is intractable when the number of resources is large. We study a family of re-solving heuristics that periodically re-optimize an approximation to the original problem known as the deterministic linear program (DLP), where random customer arrivals are replaced by their expectations. We find that, in general, frequently re-solving the DLP produces the same order of revenue loss as one would get without re-solving, which scales as the square root of the time horizon length and resource capacities. By re-solving the DLP at a few selected points in time and applying thresholds to the customer acceptance probabilities, we design a new re-solving heuristic with revenue loss that is uniformly bounded by a constant that is independent of the time horizon and resource capacities. This paper was accepted by Kalyan Talluri, revenue management and market analytics.

When Demands Evolve Larger and Noisier: Learning and Earning in a Growing Environment

Many emerging businesses and services encounter growing environments where the stochastic demand gradually increases. The increases are often not only in the expectation of the demand but also in the variance. Because of the growing environments, sequential decision making problems raise additional challenges when model parameters are unknown and need to be dynamically learned from sequentially observed data. In this work, we use a single-product dynamic pricing problem to illustrate how the non-stationary growing environment influences policy design and policy performance such as regret. We prove matching upper and lower bounds on regret and design near-optimal pricing policies. We then demonstrate how the growth rate of demand variance affects the best achievable policy performance as well as the near-optimal policy design. In the analysis, we also prove that that whether the seller knows the length of time horizon in advance or not render different optimal regret orders.

Determining an optimal replenishment run time and lot size for an imperfect manufacturing system and rework under markdown policy

This study proposes a continuous production inventory model for deteriorating items with preservation technology, rework and price-dependent demand under markdown policy. In this study, we have adopted the markdown policy to reduce inventory cost and clearance at the end of the season or to sell old or slow-moving inventory at the end of its life. Rework is another main factor to reduce production cost and environmental problem. The preservation technology is also considered during product deterioration period to control the deterioration rate of products. The shortages are not allowed. We have also considered that in one cycle, the production system can produce items in several determined replenishment cycles under markdown policy. The aim of this research is to derive the optimal production-run time, the optimal replenishment cycle length, the optimal length of time horizon and the optimal production quantity so as to minimize the total average cost. An example and sensitivity analysis is shown to illustrate the model.

Integrate carbon dynamic models in analyzing carbon sequestration impact of forest biomass harvest

Biomass is an attractive natural energy resource for mitigating climate change. However, the loss of carbon sequestration as an ecosystem service due to biomass harvest has not been considered in previous studies. To assess the impact of biomass harvest on carbon sequestration, carbon dynamics in the forests and the atmosphere were integrated. The impact of forest biomass harvests on carbon sequestration was assessed based on the difference between carbon sequestration after harvest and carbon sequestration without harvest. A Chapman-Richards function and the forest vegetation simulator (FVS) were used to simulate the growth of a forest stand. The carbon dynamics in the atmosphere were simulated by the Bern2.5CC carbon cycle model. Characterization factors of the impact were calculated in three time horizons: 20-, 100- and 500-year. According to the simulations, postponement of harvest and low harvest intensity could prolong the compensation period. The annual impact on carbon sequestration was mostly negative over a short time and became positive in the end of compensation period. The highest characteristic factors of the impact on carbon sequestration were found in rotation length of 100years with the time horizon of 500-year in the Chapman-Richards simulation and in the lowest harvest intensity with the time horizon of 500-year in the FVS simulation. Based on the results, increasing growth rate, postponing harvest, reducing harvest intensity and increasing length of time horizon could reduce the impact of forest harvest on carbon sequestration. The method proposed in this study is more proper to assess the impact on carbon sequestration, and it has much wider applications in forest management practice.

The cost-utility of aflibercept for the treatment of age-related macular degeneration compared to bevacizumab and ranibizumab and the influence of model parameters.

Age-related macular degeneration (AMD) is a blinding disease placing considerable burden on society due to blindness-associated costs. Intravitreal anti-vascular endothelial growth factors (anti-VEGFs) are effective in reducing the incidence of blindness, but at potentially high costs, depending on the cost of the drug used. Aflibercept has been introduced as an anti-VEGF equally effective to ranibizumab, but less costly. For this new drug, new cost-effectiveness analyses are needed, and AMD models used today give biased results. We investigated the cost-effectiveness of aflibercept compared to bevacizumab, ranibizumab, and no treatment and studied the influence of commonly used model parameters. A patient-level, visual acuity-based, 2-eye model was developed. Data on effectiveness were derived from randomized controlled trials evaluating the outcomes of aflibercept, bevacizumab, and ranibizumab. Utility and resource utilization were assessed in interviews with AMD patients. Costs were based on standard health care cost prices. Time horizons were two and five years. A societal perspective was employed. Over five years, costs associated with aflibercept treatment were <euro>36,030, with 2.15 QALYs. Costs associated with the bevacizumab regimens, ABC study as-needed (PRN); CATT study PRN; and CATT study 1×/month, were <euro>19,367; <euro>26,746; and <euro>30,520, with 2.16; 2.17; and 2.15 QALYs, respectively. Costs associated with ranibizumab PRN and 1×/month were <euro>45,491 and <euro>74,837 with 2.16 and 2.15 QALYs, respectively. 'No treatment' was associated with <euro>9530 and 1.96 QALYs. The incremental cost-effectiveness ratios versus 'no treatment' were: aflibercept-<euro>140,274; bevacizumab-<euro>51,062 (ABC PRN), <euro>83,256 (CATT PRN) and <euro>110,361 (1×/month); ranibizumab-<euro>181,667 (PRN) and <euro>349,773 (1×/month). Results were highly dependent on whether only one or both eyes were included, length of time horizon, and whether the costs of blindness and low-vision were included in the analysis. Aflibercept is a cost-effective treatment for AMD over ranibizumab. However, aflibercept is not a cost-effective treatment when compared to bevacizumab. Application of inappropriate model assumptions leads to a biased cost-saving estimate of the cost-effectiveness of aflibercept. Therefore, cost-effectiveness analyses should be conducted with appropriate models.

Computing a Classic Index for Finite-Horizon Bandits

This paper considers the efficient exact computation of the counterpart of the Gittins index for a finite-horizon discrete-state bandit, which measures for each initial state the average productivity, given by the maximum ratio of expected total discounted reward earned to expected total discounted time expended that can be achieved through a number of successive plays stopping by the given horizon. Besides characterizing optimal policies for the finite-horizon one-armed bandit problem, such an index provides a suboptimal heuristic index rule for the intractable finite-horizon multiarmed bandit problem, which represents the natural extension of the Gittins index rule (optimal in the infinite-horizon case). Although such a finite-horizon index was introduced in classic work in the 1950s, investigation of its efficient exact computation has received scant attention. This paper introduces a recursive adaptive-greedy algorithm using only arithmetic operations that computes the index in (pseudo-)polynomial time in the problem parameters (number of project states and time horizon length). In the special case of a project with limited transitions per state, the complexity is either reduced or depends only on the length of the time horizon. The proposed algorithm is benchmarked in a computational study against the conventional calibration method.

Case Study: Solving a Rental Fleet Sizing Model with a Large Time–Space Network

This article addresses a rental fleet sizing problem (RFS) in the context of the truck rental industry, subject to uncertain customer travel time and nonstationary customer demand that is dependent on geographical location, time, and the economic cycle of the industry. We integrate tactical (asset purchases and sales) and operational (empty truck movement and vehicle assignment) decisions, with the explicit incorporation of an asset age factor, to achieve lower cost solutions. Typically, the length of time horizon and number of locations under consideration are quite large, which makes the RFS model computationally challenging to solve. Aggregation procedures are employed for location clustering and end-of-horizon effects are examined through demand scenario-based analyses. For the reduced time–space networks, decision analyses are conducted for the RFS model to provide insights into the truck rental business regarding asset movement decisions and asset procurement/disposal decisions over time and locations.

Equilibrium Correlations of Asset Price and Return

Two empirical questions concerning the equity and housing have been studied extensively: (1) Are the price and return serially correlated, and (2) What is the optimal weight of housing in the portfolio? The answer to the second question crucially depends on the cross-correlation of assets. This paper complements the literature by building a simple dynamic general equilibrium with fully rational agents, and obtain closed form solutions for the implied auto- and cross-correlations. The length of time horizon, as well as the persistence of economic shock matter. Implications and future research directions are then discussed.

Determining optimal road class and road deactivation strategies using dynamic programming

Forest managers are faced with complicated road construction and deactivation decisions. When construction, upgrading, and deactivation strategies must be determined simultaneously over broad spatial and temporal scales, the problem becomes very complex and decision support systems are needed. In this paper, we report the development and application of an optimal road class and deactivation model using dynamic programming. We tested our model on projected road networks on Hardwicke Island, British Columbia. Sensitivity of inputs such as construction costs, upgrade costs, hauling and maintenance costs, deactivation costs, length of time horizon, discount rate, and haul volume were tested within and between two road networks. Comparison of road networks revealed that haul volume concentration, average haul distance, and total road length are the most important variables that affect road class decisions and total network costs. Within our case study, the road network with the lowest average hauling distance resulted in the lowest total cost (CAN$0.24/m3 less), because hauling costs are the largest component (46%) of total transportation costs. The dynamic programming model can be used to assess numerous road construction and maintenance assumptions under various silviculture and harvest systems.