BY MICHAEL S. GODOROV AND LOUISE M. GROSS, PPL ELECTRIC UTILITIES, AND ALVIN JACKSON, ACLARA
PPL Electric Utilities implemented a meter data management (MDM) system in 2006 with customer-facing and operational applications that has provided significant benefits. This implementation required a quality project management effort that involved numerous utility business stakeholders. Although there are many elements in a project engagement, stakeholder involvement ensured success.
An MDM system can have many utility benefits. It starts with using valued data from an advanced metering infrastructure (AMI) to improve operational efficiency, revenue management and load research and settlement. An MDM establishes the infrastructure required to transform data into information. In addition, it can be the key to informing and motivating utility consumers through bill analysis and usage information, proactive alerts, demand response and energy adviser tools. These benefits can be realized only through quality project management engagement, well-structured deployment and an effective transition to utility operations.
Effective project management involves a four-pronged effort to ensure the challenges are addressed and best practices and proper tools are applied. This effort (see Figure 1) requires choosing the best project manager through assessment of personality, strengths, weaknesses and project management experience. This experience includes how well the candidates have established strong project teams, managed customer expectations and how effectively they managed concerns and issues. Interviewing potential project managers in the project preparation phase can prove beneficial.
Staying the course is important to keep within scope, time and budget. These elements ensure project success through creation of realistic project plans, tracking project margins, mitigating risks and clearly communicating dates and responsibilities to project team members and involved business units.
An effective engagement involves multiple components across project team members. Early stakeholder and end user involvement with the core project team creates buy in and a sense of ownership and accountability. In addition, stakeholder engagement provides governance that is necessary for issue escalation and resolution at the right levels within the project organization. As a result, it enables the building of interdependencies in business relationships to ensure success.
Once the team is established, regular communication is a must. Successful communication through daily team meetings, weekly project manager meetings and executive updates facilitate trust in a project’s ability to meet stakeholder needs. It is better to overcommunicate than not at all.
Project management challenges, best practices
PPL Electric Utilities’ project architecture (see Figure 2) facilitated communication to all levels of responsibility within the project. As the first to implement an MDM of this magnitude, this structure worked well.
When embarking on an MDM deployment, understanding the business drivers and needs help set clear requirements (see Figure 3). Engagement of the stakeholders in setting the requirements will establish the criteria for success that is integral to reaching acceptance. For example, a joint effort is required to achieve a clear vision on the to-be processes while identifying and comparing them to the as-is process. This effort enables improvement of existing processes and the ability to define new ones. In addition, this type of effort leads to acceptance of some requirements that will be set by the process improvement process.
In performance, the system’s database must be sized adequately before purchasing hardware. Along with the sizing, the data dependencies must be understood, as well as when to evaluate the data generation triggers so smooth performance occurs throughout the project and in preparation for operational transition. Through the deployment, keep the business owner involved in system testing and application. This entails establishing a joint testing strategy with the dependencies defined, as well as integration testing prior to user acceptance. Involving end users in testing is beneficial twofold: hands-on training plus the ability to step into the transition from project to operations.
Deployment Challenges, best practices
Testing is essential for the project and ongoing operations. At first, PPL operated with a production database and test database. The production database was dynamic and the test database was stagnant. This option created unique data conditions and was difficult to determine if a code issue or data condition caused an anomaly.
The addition of a third development environment allows testing and the ability to verify results by comparison of the two environments. This reduced the number of issues during testing because of data conditions.
A clear statement of work generated by the vendor, business and project team aids in overall acceptance of the final product. Everyone’s adherence to this document ensures the vendor and project team meet the deliverables and the business remains within scope. To ensure acceptance, identify and prioritize all the issues based on criteria set throughout the project. Prompt attention to the issues and resolution is paramount to an amicable business relationship with the vendor.
Transition to Operations
Four critical areas ensure an effective operational transition occurs from the project to the business (see Figure 4).
Most likely, roles and responsibilities will be established to operate and maintain the new system. These roles can be composed of data analytics, database maintenance, system and application testing, as well as establishment of a system administrator. These functions are paramount in providing the necessary oversight and tuning to minimize system downtime.
Although the MDM has multiple data dependencies (i.e., meter, weather and validation, editing and estimation (VEE)), applications built on top of the data also have functional dependencies required for daily operations, such as interval billing, settlement and forecasting, and customer data presentment and analysis. A checklist that will enable the system to go live is essential in the success of the transition. A detailed schedule of when and which interfaces will run depend on one another. Assigning clear accountability to each item in the checklist provides understanding to those in the new roles and responsibilities.
Operations Challenges, best practices
Initially, PPL Electric Utilities anticipated its MDM database would grow to 2 terabytes (TB) once data was loaded and the system was being used. It quickly became apparent the capacity needed to be much greater. As the database grew, so did the operational run time. Database tuning and indexing was vital to keeping operations running smoothly. In addition, purge routines were added to limit the growth of the database. Currently three years of meter data are stored, resulting in a database size of 6 TB.
To date, tuning and performance monitoring are considered part of normal operations. Plans include the implementation of an MDM transaction mart that is anticipated to store up to seven years of meter data for use in analytics and reduce the MDM database to two years of data for business operations.
Keeping the data in sync with the core billing-customer information-distribution systems is challenging based on the number of daily transactions. One of the biggest hurdles PPL Electric experienced was the number of transactions posted to the core system for backdated scenarios. This resulted in processing errors that had to be resolved to keep the MDM in sync with the core systems in which the MDM interfaces.
Currently PPL Electric relies on the MDM for critical business processes. Therefore, emphasis is placed on regression testing. The utility approaches regression testing similar to that of a project, whereby its AMI operations group provides a high-level test plan executed by end users. AMI operations facilitates the critical path, timeline, resource requirements and vendor support. The process takes significant time, but PPL Electric expects rigorous testing and finding bugs before moving to production prevents production problems that adversely affect critical business processes across functional groups.
Unlike traditional MDMs that hold meter, weather and VEE data, PPL Electric’s MDM also has four additional applications (see Figure 5) that require 20 interfaces. These interfaces are essential to ensuring the MDM applications function nearly flawlessly for business users. Establishing these interfaces took considerable effort through development and testing by the company’s information technology, project team and stakeholders to enable data to flow from the PPL systems to the applications, especially those that were customer-facing. This effort enables PPL Electric customers to easily navigate through http://pplelectric.com to access their hourly and daily data to make decisions about their energy usage.
Deployment and operation of an MDM is a huge undertaking that requires careful planning, significant involvement from the stakeholders and business owners and commitment from the vendor to make it a success.
Louise Gross has been with PPL 27 years and is an AMI operations specialist. She is responsible for the daily operations of the MDM system. In addition, Gross supports business units at PPL by providing detailed systems analyses that enable PPL to further refine the functions of the MDM system.
Michael S. Godorov is a senior project manager at PPL Electric Utilities involved with smart grid activities. He has a Bachelor of Science in Mechanical Engineering from Spring Garden College and a Master of Business Administration from the University of Scranton. He is a registered professional engineer in Pennsylvania.
Alvin Jackson is vice president of professional services at Aclara, where he is responsible for providing direction, oversight and fiscal responsibility to delivery teams for all STAR Network (gas and water AMI), meter data management and AMI device management engagements. Jackson has a bachelor’s degree in aeronautical and astronautical engineering from the University of Washington in Seattle and a master’s degree in aeronautical and astronautical engineering from Stanford University.
PowerGrid International Articles Archives
View Power Generation Articles on PennEnergy.com