Table of contents(19 chapters)
This chapter examines the use of mathematical programming to remove systematic bias from demand forecasts. A debiasing methodology is developed and applied to demand data from an actual service operation. The accuracy of the proposed methodology is compared to the accuracy of a well-known approach that utilizes ordinary least squares regression. Results indicate that the proposed method outperforms the least squares approach.
The relationship between electricity demand and weather in the United States has been studied as of late due to increased demand, de-regulation, and new pricing models. The influence of weather or seasonality in energy consumption, particularly electricity demand, has been widely researched. A significant scientific interest in the seasonality of energy consumption has led to an important number of papers exploring the role of weather variability and change on energy consumption. Most of these papers model demand as a function of seasonal climate factors.
The goal of this research is a broad examination of monthly residential electricity demand for a region of the mid-Atlantic using Excel and step-wise regression. This is achieved by using a sequence of models built in Excel in which different patterns are gradually introduced in the estimations. Data over a seven-year period is utilized. A backward elimination step-wise regression analysis is employed to determine which independent variables best model the data. Initial independent variables included high monthly temperature, low monthly temperature, time, year, month, seasonal quarter, and introduction of a “green” tax credit for solar and wind energy.
Models for forecasting the electricity demand and the predictive power of these models is assessed. The work is organized as follows: Data description and the methodology, trend and the seasonality of electricity usage in the mid-Atlantic region, the predictive power and seasonality of the models, and main conclusions drawn from the study.
The challenges that propane companies face in maintaining a balance in inventories during the summer and winter months, and the factors that influence the residential propane demand were addressed. This chapter presents a forecasting model for propane consumption within the residential sector. Forecasting the propane demand helps to determine whether there will be a shortage of propane in the storage or distribution center, and is there a need for new distribution station or a storage facility, or vice-versa that there is an overabundance of propane, that is, far more than the demand and if there is a need to shut down few facilities. The dynamic behavior of different variables that affected the propane consumption was studied and using Base SAS we developed a forecasting model. The results indicated that the forecasting model provides a potentially useful forecast for residential propane consumption. This research has been limited to forecasting for normal periods, that is periods without irregularities in demand caused by holidays or festivals. The forecasts developed were useful in improving the inventory balance for a local propane company during different months.
This chapter considers warranty policies involving two attributes, such as the time elapsed since sale of the product and product usage at a given point in time. Examples of such policies are found for automobiles, where warranty may be invoked by the consumer if both time and usage are within specified warranty parameters when a product failure occurs. Here, we assume that usage and product age are related through a random variable, the usage rate, which may have a certain probabilistic distribution as influenced by consumer behavior patterns. Furthermore, product failure rate is influenced by the usage rate and product age as well as research and development expenditures per unit. It is assumed that, in production, there is a learning effect with time. The attained market share of a product will be influenced by the warranty policy parameters of warranty time and usage limit and also by the product price and product quality. An integrated model is developed to address multiobjective goals such as attainment of a specified level of market share and net profit per unit when manufacturing and warranty costs are taken into account. The impact of the goal priorities are investigated on the attained warranty policy parameters.
We propose a method for forecasting bank solvency that quantifies bank solvency as the probability that a bank will have more than 0.25 of the cash to total asset ratio. Predictor variables include the ratio of loans secured by farmland to total loans, the ratio of loans to farmers to total loans, and the ratio of commercial and industrial loans to total loans. Loans secured by farmland to total loans significantly predicted the potential for insolvency. To a secondary extent, commercial and industrial loans significantly predicted bank failure. This result was validated with predicted probabilities significantly explaining cash to total assets.
We evaluate the performance of financial analysts versus naïve models in making long-term earnings forecasts. Long-term earnings forecasts are generally defined as third-, fourth-, and fifth-year earnings forecasts. We find that for the fourth and fifth years, analysts' forecasts are no more accurate than naïve random walk (RW) forecasts or naïve RW with economic growth forecasts. Furthermore, naïve model forecasts contain a large amount of incremental information over analysts' long-term forecasts in explaining future actual earnings. Tests based on subsamples show that the performance of analysts' long-term forecasts declines relative to naïve model forecasts for firms with high past earnings growth and low analyst coverage. Furthermore, a model that combines a naïve benchmark (last year's earnings) with the analyst long-term earnings growth forecast does not perform better than analysts' forecasts or naïve model forecasts. Our findings suggest that analysts' long-term earnings forecasts should be used with caution by researchers and practitioners. Also, when analysts' earnings forecasts are unavailable, naïve model earnings forecasts may be sufficient for measuring long-term earnings expectations.
The aim of this research is to develop a model to forecast short-term health cost changes. The motivation for producing such a model is to provide local decision makers with a tool to predict short-term health-care costs in their localities. In order to achieve this objective, we collected data on total health-care expenditures and demographic data for California counties from 2000 to 2007. We then used various statistical methods to better understand the data and developed a regression model. Each year's prediction model was then used to forecast the following year's total health-care expenditure. The model developed adequately predicted health-care costs for the years on which the model was developed (2000–2006), and adequately forecast health-care costs for the holdout year, 2007. The average adjusted R2 value was 0.57, with an average mean absolute deviation score of 34. The best predictors of total health-care expenditures were county population, the number of county health-care facilities, and county per capita personal income. The practical implications of the model are that it will provide public and private decision makers with a useful tool for forecasting short-term demand for health-care services, enabling better planning for health-care manpower, facility planning, and financial planning needs. The contribution of this paper contrasts with the earlier work in that it supports short-term operational, not strategic, planning needs. The paper's limitation is that it relies on data from one state. It should be tested in other, dissimilar, areas of the United States.
Hospitals provide care to several thousand patients hospitalized in various nursing units. This process involves admissions of patients entering the nursing units and discharges of patients leaving the nursing units. These admissions and discharges have been identified as hand-off points, and such transitional points have higher potential for errors. Furthermore, admissions and discharges also have several associated activities that need to be accomplished, thus causing an increase in the need for resources like nursing staff, etc., which impacts on efficiency. Hence, a better understanding of the trends and patterns in admissions and discharges is necessary to improve the safety and efficiency of healthcare processes. This healthcare-related forecasting case study uses admission and discharge data from more than 100 thousand patients from 38 nursing units like medicine, surgery, step-down, pediatric, etc., over a three-year period from October 2007 through October 2010 in a large academic health system in the United States. There are two primary goals for this study: (1) to perform pattern analysis on the admission and discharge data, for facility and workforce planning and determining shift structure purposes and (2) to perform forecasting for corrective allocation purposes. Similar methods can be used by other hospitals to improve safety and efficiency.
The authors have previously validated a design of the health-care supply chain which treats patients as inventory without loss of respect for the patients. This work continues examination of patients as inventory while addressing the dual objectives of reducing redundancy in services and creating greater efficiency in the health-care supply chain. Historical data is used to forecast health care needs in light of the increasingly specialized health-care professionals, which have resulted in much more flexible and expensive supply chains. The lack of common data storage, or electronic medical records (EMRs), has created a need for redundancy (or rework) in medical testing. The use of EMR will also enhance our ability to forecast needs in the future. We perform simulations using SigmaFlow software to address our goals relative to the resource constraints, monetary constraints, and the overall culture of the medical supply chain. The simulation outcomes lead us to recommendations for data warehousing as well as providing mechanisms, like inventory postponement strategies, to establish structures for more efficiency, and reduced flexibility in the supply chains.
This chapter describes the method we use to predict the demand for a new brand over its entire future life cycle before the brand is introduced into the market. The forecast is a prior forecast prepared at time t−1 using all the data that is available up till time t−1. Additional information about the future in the form of advance orders becomes available at time t. The advance orders contain the customers' plans for future purchases. They contain therefore a forecast for future demand. This chapter discusses how at time t−1 the prior forecast of, and the estimate of the locus of, a new brand's life cycle (based on information up to t−1) for the future periods t+1, t+2, …,t+k are developed. The chapter discusses how at time t when advance orders become available for the future periods t+1, t+2, …,t+k the prior forecast is updated for the length of the life cycle with this new information. These updates are made using Kalman's filter. Using this method we have been able to obtain good estimates of the locus of the life cycle of new brands. We have also been able to predict the turning points in the brand's life cycle six months before it occurs. The chapter shows a method for developing sequentially improved forecasts.
Internal prediction markets draw on the wisdom of crowds, gathering knowledge from a broad range of information sources and embedding that knowledge in the stock price. This chapter examines the use of internal prediction markets as a forecasting tool, including as a stand-alone, and as a supplement to forecasting tools. In addition, this chapter examines internal prediction market applications used in real-world settings and issues associated with the accuracy of internal prediction markets.
Increasing competition within the global supply chain network has been pressuring managers to improve efficiencies of production systems while, at the same time, reduce manufacturing operation expenses. One well-known approach is to have better control of the manufacturing system through more accurate forecasting and efficient control. In other words, a production control paradigm with more reliable forward visibility should help in maintaining a cost-effective yet lean manufacturing environment. Hence, this study proposes a predictive decision support system for controlling and managing complex production environments and demonstrates a Visual Interactive Simulation (VIS) framework for forecasting system performances given a designated set of production control parameters. The VIS framework is applied to a real-world manufacturing system in which the primary objective is to minimize total production while maintaining consistently high throughput and controlling work-in-process level. Through this case study, we demonstrate the use and validate the effectiveness of VIS in optimization and prediction of the examined production system. Results show that the predictive VIS framework leads to better and more reliable decision making on selection of control parameters for the manufacturing system under study. Statistical analyses are incorporated to further strengthen the VIS decision-making process.
The accuracy of forecasts has a critical impact on an organization. A new, practical, and meaningful forecast performance measure, percentage forecasting error (PFE), was introduced by the authors in an earlier publication. In this chapter, we examined the accuracy of the PFE under several different scenarios and found the results to indicate that PFE offers forecasters an accurate and practical alternative to assess forecast accuracy.