Developing a New Long-Term Demand Forecasting Model for Seattle Public Utilities

Seattle Public Utilities (SPU) provides drinking water to 1.5 million people in the Seattle metropolitan area from two Cascade Mountain reservoirs. SPU sells approximately half of its water supply directly to retail customers and half to 24 wholesale water customers (regional cities and water districts in the region). Long-term water demand forecasting is a critical tool in ensuring an adequate water supply for metropolitan areas experiencing significant changes in customer growth and behavior. Seattle, and the Puget Sound region in general, has been one of the fastest growing parts of the U.S. for decades. Development of a forecasting tool will depend on how the utility intends to use it, the data available to build it, and understanding the drivers of long-term demand. This presentation will discuss the process of developing a new long-range water demand model for Seattle Public Utilities that informs key planning and regulatory needs through a balance of analysis and informed estimates. SPU currently uses a demand forecasting model that was developed in 2006 in support of its 2007 Water System Plan. The modeled forecast is reported to State agencies as a regulatory requirement, is used for water supply planning, conservation program evaluation, wholesale customer contract development, and retail rate setting. The 2006 model uses a Variable Flow Factor (VFF) approach, which is more complex than a trend analysis but more simplified than an econometric model. The current model forecasts water demand by sector (single and multi-family residential, non-residential) for Seattle and each of its individual wholesale customers at an annual timestep, using water demand factors known to drive consumption. SPU desired to develop a new long-term demand model to be used for the same purposes and the same level of detail, but that used more data and new techniques to achieve the same key objectives. The presentation will present findings from a literature review to explore demand modeling practices that are used throughout the industry and have been explored by academic professionals around the world. This background information will introduce the main alternative methods available including qualitative methods, time series, VFF, and econometric models and their advantages and limitations. The literature review was also used to review independent and dependent variables that can be included in model development. These variables frequently include income, weather and seasonal factors, household structure and size, property characteristics, and conservation among others. SPU chose to explore improving its variable flow factor approach with econometric modeling techniques. This process began with exploring over 30 years of billing, climate, socioeconomic, and conservation data. After a long and careful data quality assurance process, the data was used to inform the variable selection process. Selecting variables to be included in the demand forecast model is an important task. SPU's new econometric approach to model development allowed for a more data-intensive approach to variable selection. Variable selection was achieved by examining combinations of variables and assessing their performance with several statistical measures. in order to measure factors such as their overall explanatory value, the reasonableness of the influence statistically attributed to each variable and the variables' relationship to one another (correlations and collinearity). The SPU team learned many lessons about how to identify and define variables, data sources, and processing requirements. We will discuss these lessons and present the variables that were selected to be used in the updated long-term forecast. We will discuss the process that was undertaken to explore and evaluate multiple econometric models, the final forecast equation selection process, and implementation of the model into a user-friendly MS Excel spreadsheet tool. SPU anticipates this long-term demand model to be a tool that is regularly used, updated, and enhanced to ensure that best practices are incorporated throughout.