Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Business Analytics & Data Mining
Transcript of Business Analytics & Data Mining
Industry: Energy utilities
is the process of iterative, methodical exploration of an organization's data with emphasis on statistical analysis.
Focus on Data-Driven Decision making.
Performed in four types or steps, to answer a series of business-driven questions:
"What happened in the past?"
"Why did it happen?"
"What is likely to happen?"
"What should I do about it?"
Example of Analytics in action: Bar deciding a happy hour special.
The process of analyzing raw data in a number of different ways and dimensions, and summarizing and condensing it into actionable information by looking for clusters, associations or patterns.
Also known as data discovery or knowledge discovery.
Six phases of Data mining:
Identify objectives, issues, and requirements
Examine metadata for quality and relevance
Clean and format data in preparation for modeling
Apply mining functions to data to organize into useful info
Evaluate the model to see if optimal results have been obtained
Export data into tables, reports, graphs, etc.
Founded in late 1800's as subsidiary of the Washington Traction and Electric Company
Divested itself of streetcar operations in 1935, becoming the Potomac Electric Power Company
Reorganized as a unit of Pepco Holdings, Inc. in 2001
Provides power for nearly 800,000 customers in D.C. and Maryland
Utilities are diversifying their portfolio by investing in distributed renewable generations like wind and solar.
The distribution grid is becoming more dynamic as new technologies are gaining ground. E.g.: customer-owned distributed resources, energy storage, smart metering & electric vehicles charging stations.
Aging infrastructure, unusual weather patterns, cyber security and data privacy threats are challenging the utility's ability to maintain reliability of service.
The installation of new "smart" technologies have provided utilities with tons of new data and made existing data more accessible. What to do with all this data?
Today's consumers expect a rich, tailored "customer experience."
The grid will become more dynamic with increase in renewables like wind and solar, which come on and off the grid. EV charging will become more popular.
As penetration of new technologies grows, there will be an impact on operations, commercial relationships, and distribution infrastructure.
Automated demand response is fast becoming a necessity which, in future, will lead to the development of a distribution marketplace.
Business Analytics & Data Mining will be needed to support this type of market.
Improve their operations and customer satisfaction by:
Protecting revenue (customer)
Improving customer engagement (customer)
Maintaining reliability and securing the grid (operational)
Reducing the cost of operations (operational)
Complying with environmental regulations (operational and customer)
Take advantage of the data to create value for their business by building:
Infrastructure (servers, storage, clustering software, networks, etc.),
Data collection from 'smart' devices (streaming data, monitoring and analysis),
Data management (software that cleanses, normalizes, tags, and integrates data),
Analytics & business intelligence (real-time analysis & automated rules-based transactional decision making),
Decision support (collaboration, scenario evaluation, risk management, and decision capture and retention).
Smart grid, smart network devices, SCADA, RTUs, power line sensors, etc are generating more data than ever before.
The bulk of the data that Pepco collects through smart meters and grid-generated data, is structured data. It is time series data recorded at uniform time intervals ranging from seconds to hours.
The data collected by the smart meter is collected only periodically. To support the dynamic distribution network of the future, high-velocity data discovery based on interactions between utilities and smart homes and buildings will be required.
Financial benefit comes from the ability to filter through large amounts of data, unlock insights available through new data sources, and leverage and optimize untapped data.
Target marketing, evaluate effectiveness of demand response & energy efficiency programs, demand response incentives, analytical tools to help customers reduce their energy bills, communications about outages.
Pinpoint location of outages, scan for potential security breaches, perform analysis of faults or voltage irregularities.
Detect fraud and theft, identify unbilled accounts.
Predictive asset management:
Forecast potential performance or equipment failures by studying load on each asset and rated capacities of devices.
Capital investment planning:
Formulate capital investment strategy by accessing timely information about grid and asset status.
Lack of awareness
Lack of competition
Uncertainty about costs and requirements
Recognize the implications of operating without critical information.
Conduct detailed analysis to determine what new technology and staff investments are required.
Consider cloud services and a shared services model with other utilities to reduce costs.
Recognize value of untapped data in supporting fact-based decisions in any industry. But by itself, this wealth of raw data is useless!
Utilizing analytics and data mining, managers can take esoteric data and transform it into meaningful, actionable information.
Formulate a strategy that includes evaluation of decision makers' requirements, decision processes, new technology, legacy systems, and availability & quality of data.
Presented by: Deotima Gangopadhyay & Edward Rowe