Stochastic Forecasts Research Articles

Theoretical development of stochastic delay analysis and forecast method

Schedule delay analysis and claim management for construction projects have received much attention recently due to an increasing number of construction disputes. Most of the available delay analysis methods employ a deterministic approach that uses deterministic activity duration in as-planned schedules for schedule comparison. However, the duration of construction projects is naturally uncertain. Based on a mathematical-model delay analysis method, this study proposes an innovative approach, the Stochastic Delay Analysis and Forecast method (the SDAF method), which combines probabilistic activity duration and systematic delay analysis procedures to predict the contribution of a single activity’s delay and its probability to cause overall project delay. This paper proposes a methodology and examines its results by means of a simulated case. Although this study has simplified practical problems, preliminary study results are valuable and form a basis for further research. Namely, these study results provide not only a stochastic delay analysis and forecast method but also a fundamental basis for developing advanced schedule delay analysis methodology to help in solving complicated schedule delay problems in construction projects.

Simulation based comparison of stochastic claims reserving models in general insurance

The appropriate estimation of incurred but not reported (IBNR) reserves is traditionally one of the most important task for property and casualty actuaries. As certain claims are reported considerably later after their occurrence, the amount and appropriateness of the reserves have a substantial effect on the financial results of institutions. In recent years, stochastic reserving methods have become increasingly widespread, supported by broad actuarial literature, describing development models and evaluation techniques. The cardinal aim of the present paper is to compare the appropriateness of several stochastic estimation methods, supposing different distributional underlying development models. For lack of analytical formulae in most of the model settings relevant from a practical perspective, due to the complex behavior of summed variables, simulations are performed to approximate distributions and results. Considering that the number of runoff triangles is generally limited, stochastically simulated scenarios contribute to feasible solutions. Stochastic reserving is taken into account as a stochastic forecast, thus comparison techniques developed for stochastic forecasts can be applied, opening up new informative perspectives beyond classical prediction measures, such as the mean square error of prediction.

Solving the Multivariant EV Routing Problem Incorporating V2G and G2V Options

In the near future, gasoline-fueled vehicles are expected to be replaced by electrical vehicles (EVs) to save energy and reduce carbon emissions. A large penetration of EVs threatens the stability of the electric grid but also provides a potential for grid ancillary services, which strengthens the grid, if well managed. This paper incorporates grid-to-vehicle (G2V) and vehicle-to-grid (V2G) options in the travel path of logistics sector EVs. The paper offers a complete solution methodology to the multivariant EV routing problem rather than considering only one or two variants of the problem like in previous research. The variants considered include a stochastic environment, multiple dispatchers, time window constraints, simultaneous and nonsimultaneous pickup and delivery, and G2V and V2G service options. Stochastic demand forecasts of the G2V and V2G services at charging stations are modeled using hidden Markov model. The developed solver is based on a modified custom genetic algorithm incorporated with embedded Markov decision process and trust region optimization methods. An agent-based communication architecture is adopted to ensure peer-to-peer correspondence capability of the EV, customer, charging station, and dispatcher entities. The results indicate that optimal route for EVs can be achieved while satisfying all constraints and providing V2G ancillary grid service.

Forecasting the distribution of precipitate diameters in the presence of changes in the structure of the material

AbstractThe results of investigations on the microstructure of T23 and T24 low-alloy steels as well as P91 and P92 high-chromium steels in the as-received condition and after 70.000 h annealing at 550-650°C are presented. The quantitative analysis of the existing precipitates was performed for representative images of microstructure. The statistical analysis of collected data allowed the parameters of a selected theoretical statistical distribution to be estimated. A forecast of average precipitate diameter and standard deviation of such a distribution for the time of 100,000 hours at 550 and 600°C for T23 and T24 steels and at 600 and 650°C for P91 and P92 steels was calculated. The obtained results of investigations have made it possible to compare changes in the microstructure of various steel grades due to long-term impact of elevated temperature. They have also confirmed the possibility of using, in evaluating the degradation degree of materials in use, the forecasting methods that derive from mathematical statistics, in particular the theory of stochastic processes and forecast by analogy methods. The presented approach allows the development of a forecast of precipitate diameter probability density under the microstructure instability conditions for selected steel grades. The assessment of material condition that takes into consideration, but is not limited to, the precipitate diameter measurement is useful as an assessment component in estimating the time of safe service of power unit elements working under creep conditions.

Handling Risk and Uncertainty in Portfolio Production Forecasting

Summary Stochastic modeling provides a mechanism for incorporating risk and uncertainty considerations into portfolio production forecasts. Through this process, insight is gained into the likelihood of production targets being missed, met, or exceeded. This insight enables organizations to better manage operational, positioning, and strategic planning activities around stakeholders’ production expectations. Inherent in all capital programs are numerous uncontrollable, but definable, factors that affect overall corporate production performance. These factors can be categorized into four groups: (1) timing uncertainties, (2) performance uncertainties, (3) sequencing uncertainties, and (4) risk. Timing uncertainty considers spud scheduling, spud-to-first-production cycle timing, and production-ramp-up cycle timing. Performance uncertainty considers the historical or modeled distribution of period-specific production rates within the constituent plays (e.g., What is the unavoidable range of variability within a play as depicted in a peak-normalized composite-production plot of wells within the analog population?). Sequencing uncertainty considers performance-percentile clustering or sequencing within the program (e.g., the number of top-quartile wells that are, by chance, drilled early in the year vs. later in the year). Finally, risk addresses commercial failure within a program attributed to either geology or execution, or both. By integrating historical operational data with a standardized set of play-assessment deliverables, the building blocks of a stochastic capital-program forecast and analysis are readily available. Ultimately, the use of stochastic modeling in portfolio production forecasting provides an organization's decision makers with the information necessary to examine investment and strategic decisions in the context of corporate-risk tolerance.

Stochastic Population Forecast for Germany and Its Consequence for the German Pension System

Population forecasts are crucial for many social, political and economic decisions. Official population projections rely in general on deterministic models which use different scenarios for future vital rates to indicate uncertainty. However, this technique shows substantial weak points such as assuming absolute correlations between the demographic components. In this paper, we argue that a stochastic projection alternative, with no a priori assumptions provides point forecasts and probabilistic prediction intervals for demographic parameters in addition. Age-sex specific population forecast for Germany is derived through a stochastic population renewal process using forecasts of mortality, fertility and migration. Time series models with demographic restrictions are used to describe immigration, emigration and time varying indices of mortality and fertility rates. These models are then used in the simulation of future vital rates to obtain age-specific population forecast using the cohort-component method. The consequence for the German pension system is discussed. To maintain the actual average pension level the premium rate of the present system rises at least by 50% as the old-age ratio nearly doubles by 2040.

A Novel Methodology of Stochastic Short Term Forecasting of Cloud Boundaries

Following the chaos theory proposed by Lorenz, probabilistic approaches have been widely used in numerical weather prediction. This paper introduces a convection diffusion equation to quantify the dynamic growth of uncertainty for short term forecasts of cloud boundaries. The equation is inserted into a numerical weather prediction model, weather research and forecast. A two parameter model based on wind velocity dispersion and surface evaporation rates parameterizes the stochastically motivated, but deterministic, equation. Prediction verification tests in comparison to observed data show good predictive capability for an hour, with a gradual loss of predictive power continuing for predictions up to three hours shown qualitatively. The methodology can be applied to a variety of topics in numerical weather prediction research.

Stochastic Forecasting of Vehicle Dynamics Using Sequential Monte Carlo Simulation

Predictive control is a popular approach for further improving the efficiency and performance of vehicular systems enabling intelligent systems behavior appropriate to the driving situation. To calculate such control strategies, the future vehicle dynamics or subsequent states have to be predicted. We introduce a stochastic framework based on an explanatory model and stochastic processes to predict future vehicle dynamics with road network data. The distributions of the future states are approximated using sequential Monte Carlo simulation. The proposed approach enables stochastic forecasts incorporating uncertain driver behavior and available road data. Parameter inference is shown for exemplary real-drive test data, and predictive performance is evaluated using commonly used reference models. The results show that the explanatory model provides more specific information than time-series models do, still considering the uncertainty in the driver's behavior or the situation. The framework can be applied with predictive control algorithms enabling intelligent control of vehicular systems. Furthermore, the framework or parts of it may be usable for other applications like predicting behavior of traffic participants or general characterization of driver behavior.

Accelerated Model Predictive Control for Electric Vehicle Integrated Microgrid Energy Management: A Hybrid Robust and Stochastic Approach

A microgrid with an advanced energy management approach is a feasible solution for accommodating the development of distributed generators (DGs) and electric vehicles (EVs). At the primary stage of development, the total number of EVs in a microgrid is fairly small but increases promptly. Thus, it makes most prediction models for EV charging demand difficult to apply at present. To overcome the inadaptability, a novel robust approach is proposed to handle EV charging demand predictions along with demand-side management (DSM) on the condition of satisfying each EV user’s demand. Variables with stochastic forecast models join the objective function in the form of probability-constrained scenarios. This paper proposes a scenario-based model predictive control (MPC) approach combining both robust and stochastic models to minimize the total operational cost for energy management. To overcome the concern about the convergence time increasing from the combination of scenarios, the Benders decomposition (BD) technique is further adopted to improve computational efficiency. Simulation results on a combined heat and power microgrid indicate that the proposed scenario-based MPC approach achieves a better economic performance than a traditional deterministic MPC (DMPC) approach, while ensuring EV charging demands, as well as minimizing the trade-off between optimal solutions and computing times.

The Brazilian National Institute of Meteorology (INMET) and its contributions to agrometeorology

Desde a sua criação, em 1909, no âmbito do Ministério da Agricultura, o INMET tem consciência de seu papel de provedor público de informações relacionadas a tempo, clima e áreas afins que possam ajudar o sector agrícola a fazer face aos riscos resultantes de condições atmosféricas adversas. As contribuições do Instituto desde a coleta e divulgação de dados – inicialmente obtidos apenas por meio de estações meteorológicas, mas atualmente também a partir de satélites – até produtos cada vez mais sofisticados. Estes abrangem previsões do tempo e alertas meteorológicos convencionais; mapas e dados de monitoramento do clima e climatologia; previsões numéricas de tempo detalhadas e cada vez mais precisas; modelos de previsão climática estocásticos e estatísticos, e previsões sazonais. O Instituto também dedica esforços significativos para o desenvolvimento de produtos de aplicação, especialmente para usuários do setor agrícola, de agricultores individuais e cooperativas a formuladores de políticas governamentais. O Sisdagro, um sistema de apoio à decisão para a agricultura, e o LATIS, um laboratório para a aplicação de informações de imagens de satélite e informações climáticas no monitoramento de safras, em cooperação com a CONAB, são exemplos notáveis. O INMET pretende reforçar os laços com outras instituições especializadas, nacionais e internacionais, para ampliar a gama de produtos de aplicação e serviços voltados para o setor agrícola e para melhorar seus canais de comunicação com os usuários finais, objetivando tornar mais úteis os produtos e serviços que oferece.

Four Answers to the Question: What Can I Learn From Analytics?

The future of data-driven analysis in exploration and production (E&P) will depend on whether it can add value in the field. Four examples of what is possible were presented recently at the 2016 Unconventional Resources Technology Conference (URTEC) in San Antonio with the authors of papers posing questions such as: Does it matter if a lateral is drilled toe-up or toe-down? What are the changes in a fracturing design that will offer the biggest production payoff? Why has the drilling slowdown not depressed production from unconventional gas plays? What is the half-life of my field, and why should I care? The stories below bring together advanced statistical analysis methods with multiple names: analytics, big data, machine learning, and even a “physio statistical engine for automatic stochastic production forecasting.” All this new E&P math is aimed at identifying patterns and relationships that otherwise would be missed. But automatic and self-learning does not mean all-knowing. For example, one of the programs used to analyze fracturing used facial recognition to classify pressure changes during each stage to sort them into different classes. Production data revealed that one of the classes of fractures, characterized by a pressure bump near the end of the stage that could mean trouble, were often more productive than those that went exactly according to plan. But it was up to the completion team to figure out how to apply that observation. The process is “brutally empirical,” said Roger Anderson, president of AKW Analytics, who led the study for Range Resources. He said the method is good at identifying what is happening, but it is unable to determine why or how it happened.

Looking Forward Approach in Cooperative Differential Games with Uncertain Stochastic Dynamics

In this study, a novel approach for defining and computing a solution for a differential game is presented for a case, wherein players do not have complete information about the game structure for the full time interval. At any instant in time, players have certain information about the motion equations and payoff functions for a current subinterval, and a forecast about the game structure for the rest of the time interval. The forecast is described by stochastic differential equations. The information about the game structure updates at fixed instants of time and is completely unknown in advance. A new solution is defined as a recursive combination of sets of imputations in the combined truncated subgames that are analyzed by the Looking Forward Approach. An example with a resource extraction game is presented to demonstrate a comparison of payoff functions without a forecast and that with stochastic and deterministic forecasts.

The importance of wave climate forecasting on the decision-making process for nearshore wave energy exploitation

This work presents a new methodology for the medium to long-term stochastic forecasting of the main variables and indexes related to the wave climate that are involved in the decision-making process to allocate, operate and maintain individual nearshore wave energy converters (WECs) and/or wave farms. Compared to the state-of-the-art approaches, this methodology includes the assessment of the uncertainty by means of Monte Carlo simulations, constituting a valuable step forward. The methodology is based on the simulation of Ny-year time series of wave climate variables that maintain the same statistical descriptors and seasonal and year-to-years variations of a hindcasted time series. This step is repeated Ne times to provide a sample size large enough to assess the uncertainty of the predictions. Because the wave energy resource is obtained from the nearshore, a large amount of wave propagations would be required. However, our methodology incorporates downscaling techniques that significantly improve the computational efficiency, and only a reduced number of Nw sea states should be propagated using an advanced numerical model. The methodology was applied to Playa Granada beach (southern Spain), obtaining the wave energy resource at 24 locations in the nearshore for 25-year time series repeated 1000 times. The selection of the most promising location for WECs on the basis of hindcasted or forecasted data provides different results. This highlights the importance of the proposed methodology for the advanced planning and design of any prospective energy extraction project.

Portfolio Execution with Multi-Period Stochastic Forecasts and Size Constraints

This paper investigates the problem of finding a dynamically updating trading schedule for a portfolio with stochastically evolving forecasts, absolute value based execution costs and a decaying market impact as well as size constraints. This is achieved by deriving a continuous time stochastic state evolution model as well as a Hamiltonian with a corresponding HJB equation which is then approximately solved to third order accuracy which provides a Value function and the optimal controls.

Stochastic Analysis and Forecasts of the Patterns of Speed, Acceleration, and Levels of Material Stock Accumulation in Society.

The recent acceleration of urbanization and industrialization of many parts of the developing world, most notably in Asia, has resulted in a fast-increasing demand for and accumulation of construction materials in society. Despite the importance of physical stocks in society, the empirical assessment of total material stock of buildings and infrastructure and reasons for its growth have been underexplored in the sustainability literature. We propose an innovative approach for explaining material stock dynamics in society and create a country typology for stock accumulation trajectories using the ARIMA (Autoregressive Integrated Moving Average) methodology, a stochastic approach commonly used in business studies and economics to inspect and forecast time series. This enables us to create scenarios for future demand and accumulation of building materials in society, including uncertainty estimates. We find that the so-far overlooked aspect of acceleration trends of material stock accumulation holds the key to explaining material stock growth, and that despite tremendous variability in country characteristics, stock accumulation is limited to only four archetypal growth patterns. The ability of nations to change their pattern will be a determining factor for global sustainability.

Stochastic Forecast of the Financial Sustainability of Basic Pension in China

The paper focuses on the stochastic forecast of the financial sustainability ofbasic pension, based on predictions for the population of China. The population was calculated iteratively by using Leslie matrix. An auto-regressive moving average model was adapted for the predictions of the fertility rates and the mortality rates. The Monte Carlo stochastic method was adapted for the projections of the dynamic process of the financial sustainability of the basic pension from 2013 to 2087 by 5000 times simulation. The forecasting results show that the imbalance of basic pension will occur in 2026. If the statutory retirement age is postponed by five years, the occurrence of the financial gap of the basic pension may be delayedby about 20 years, and the median deficit of basic pension will be reduced by about 64.25% in 2087.

Stochastic Monthly Rainfall Time Series Analysis, Modeling and Forecasting in Kavala City, Greece, North-Eastern Mediterranean Basin

Rainfall is one of the most important sources of water on earth supporting the existence of the majority of living organisms. Time series analysis, modeling and forecasting constitutes a tool of paramount importance with reference to a wide range of scientific purposes in meteorology (e.g. precipitation, humidity, temperature, solar radiation, floods and draughts). The present research applies the Box-Jenkins approach, employing SARIMA (Seasonal Autorregressive Integrated Moving Average) model to perform short term forecasts of monthly rainfall in Kavala city, Kavala Prefecture, Region of Eastern Macedonia-Thrace, North-Eastern Greece, North-Eastern Mediterranean Basin, modeling past rainfall time series components structure and predicting future quantities in accordance to the past. The model which is mostly fit to both describe the past rainfall data and thus generate the most reliable future forecasts is selected rated by means of both the AIC- and BIC- (SBC-) model evaluation criteria. The conclusions of this research will provide local authorities (e.g. General Secretariat for Civil Protection, European Center for Forest Fires, Deputy Governor of Agricultural Economy, daily fire risk maps designers, hydraulic, irrigation and environmental engineers, city inhabitants, farmers etc.) to develop strategic plans, policies and appropriate use of available water resources in Kavala city district.

Stochastic Forecasting of Life Expectancies Considering Multi-Maintenance Criteria and Localized Uncertainties in the Pavement-Deterioration Process

Abstract Understanding the deterioration characteristics of highway pavements is an essential factor for developing better maintenance strategies, as well as estimating mid- to long-term budget demands. In general, estimation results regarding life expectancy have often been represented with a constant in deterministic manners. However, in reality, these results do not correspond with real life spans because of the severe uncertainties in the pavement-deterioration process and various maintenance criteria. For these reasons, in this paper, the real life expectancy was assessed in a stochastic manner with the Bayesian Markov hazard model, which is suitable for estimating localized uncertainties in the progress of deterioration based on various deterioration indices. On the basis of this model, we have developed a new methodology that can determine the joint distribution of total maintenance demands by aggregating multiple deterioration processes using the concept of “safety-hazard zones.” In addition, we developed a method for disaggregating the joint distribution by maintenance types to mitigate the conventional problems associated with the probabilistic budget estimation approach. For our empirical study, we used data from the Korean national highways from 2007 to 2011 concerning pavement conditions, including crack, rut depth, and international roughness index (IRI). The results of this study are meaningful in that they establish the ideal conditions required for the Bayesian Markov hazard model to forecast the deterioration of pavement, allowing the incorporation of a practical viewpoint into an academic approach under the concept of asset management.

Macroweather precipitation variability up to global and centennial scales

AbstractWe investigate precipitation variability in the “macroweather” regime—the intermediate regime between the familiar weather and climate regimes—which is associated to time scales from about 10 days to 30–100 years. Macroweather precipitation is characterized by negative fluctuation exponents. This implies—contrary to the weather regime—that fluctuations tend to cancel each other out, they diminish with time scale, this is important for seasonal, annual, and decadal forecasts. Aiming at a wide‐scale range space‐time statistical description of macroweather precipitation, we study the scaling of three centennial, global‐scale precipitation products (one gauge based, one reanalysis based, and one satellite based) and systematically compare them over wide ranges of time and space scales. Although these products have very similar temporal statistics, at 5° resolution, they only agree with each other after being averaged over scales of several years, at scales larger than 2–3 decades, they disagree again. In space, there is less agreement on the statistics but—since the data have low resolutions (mostly 5° × 5°)—the disagreement is only over a small overall range of scales: the monthly data agree fairly well at scales 20°–30° and larger. Moreover, we quantify the outer scale limit of the temporal scaling (20–40 years, depending on the product, on the spatial scale, pixel, or global). Overall, results show that precipitation can be modeled with space‐time scaling processes. The improved understanding of the space‐time macroweather precipitation variability and the limitations of precipitation products provided by this work opens new perspectives to the stochastic modeling and forecasting of macroweather precipitation as well as separating natural and anthropogenic precipitation.

Retirement planning in the light of changing demographics

With increasing longevity and decreasing fertility rates, governments and policy makers are increasingly engaged in the question of long term retirement planning. In many cases this has included emphasising the need for individuals to take more responsibility for their own retirement planning through tax incentives, compulsion and changes to the age at which state retirement benefits become available. In the case of Australia, as is considered here, long term retirement planning has been focused around the development of a compulsory defined contribution (DC) superannuation system. Here we investigate the interaction between population ageing and the sustainability of the superannuation system by modelling a general superannuation scheme to compare the adequacy of retirement funds under a number of alternative scenarios. The model incorporates stochastic longevity forecasts and provides insight into the sufficiency of compulsory retirement saving both now and future. We find that the current pension scheme is more robust to longevity improvements for mid-class individuals however significant gaps arise for low-income individuals as longevity improves. Without addressing these issues, government expenditure is expected to increase substantially.