Bibliographic record
Abstract
Describes the forecasting process used at Columbia Gas ... accuracy of Design Day Forecasts and Daily Operational Forecasts is very critical ... both time series and regression models play an important role in forecasting. Columbia Gas of Ohio places great importance on its daily forecasting process and has spent six years developing and improving its current daily forecasting techniques. The daily demand forecasts are used in preparing strategic plans, financial projections, rate case and other regulatory proceedings, distribution system design, gas supply & capacity planning, and operational planning. Columbia's forecasting process has resulted in very small error percentages. In fact, results from a 1995 survey administered by the American Gas Association and Canadian Gas Association found Columbia's forecast accuracy to be among the best in the industry. This article discusses Columbia's forecasting process, historical accuracy of the forecast and two major types of daily forecasts developed at Columbia: Design Day Forecast and Daily Operational Forecast. Columbia Gas of Ohio (Columbia) is one of the five distribution subsidiaries of Columbia Energy Group and is the largest natural gas utility in Ohio having nearly 1.3 million customers in more than 1,000 communities. Columbia Gas of Ohio is headquartered in Columbus, Ohio. During 1999 Columbia delivered 99 BCF to Sales customers and 208 BCF to Transport customers. Columbia Energy Group, based in Herndon, Va., is one of the nation's leading energy services companies, with assets of approximately $7 billion. Its operating companies engage in virtually all phases of the natural gas business, including exploration and production, transmission, storage and distribution, as well as retail energy marketing, propane and petroleum product sales, and electric power generation. TYPES OF DAILY FORECASTS Columbia develops two types of daily forecasts: Design Day Forecast and Daily Operational Forecast. While these forecasts are used for different purposes they are developed from a single consistent forecasting process, which we will describe later. DESIGN DAY FORECAST The Design Day Forecast is primarily used to determine the amount of gas supply, transportation capacity, storage capacity and peaking contracts that Columbia needs to serve its contractually firm and human needs customers. Each year Columbia contracts for a portfolio of monthly, seasonal and annual supply contracts designed to meet the seasonal requirements and Design Day requirements for its firm customers. However, these same supply contracts must have purchasing flexibility for Columbia to respond to actual weather (warm or cold) and operating conditions. Many of the supply and capacity contracts contain a fixed cost (Demand Cost) which is paid regardless of use and a variable cost (Commodity Cost). In establishing supply and capacity levels, it is very important that the Design Day Forecast is accurate. If a company's design forecast is not accurate it may not contract for the proper supply and capacity assets. This could place the company's customers at economic or service risk. Over the past seven years Columbia's Mean Absolute Percent Error (MAPE) of the Design Day Forecast has averaged 0.4%. Table 1 shows the annual MAPE for each of the past seven Design Day Forecasts. The Design Day Forecast is based on Columbia's Design Conditions, which consist of the following: Design Current Day Temperature, Design Prior Day Temperature, and Design Wind Speed. DESIGN CONDITIONS Both Design Temperatures are developed based upon the analysis of all available historical weather data going back to 1949. Traditionally, Columbia updates this historical temperature data for analytical purposes approximately every five years. Table 2 shows the Design Temperature and Design Wind Speed for the 4 major geographical areas served by Columbia. …
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.014 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".