Pharos Enterprise Intelligence

Managing information and knowledge is an important role within an organization. Let's define some terminology. Information management is managing the process of acquiring, analyzing, disseminating, and owning information. Knowledge management is the how you record, retain, and disseminate knowledge. Knowledge is the application of information.

In managing information is it critical to have an information owner. The owner has access to the information, knows how to interpret the information and most importantly is the one to sign off on the information's use. It is not uncommon for someone to have access to information from which they run an analysis with wrong assumptions. The information could be incomplete or inappropriate. It is important for people within the organization confer with the appropriate information manager before they set out to perform an analysis. It may sound bureaucratic but it can lead to preventing costly mistakes.

In the context of the commercial operations of a power company we can divide information into different time horizons. The past, the present and the future. Examples of the past information is your settlements data and performance data. The present is your real time business intelligence such as what is my power plant doing now and what is happening in the marketplace. The future are your forecasts, forward curves and budgets.

It is important that anyone doing an analysis in any of the three horizons confers with the appropriate manager. Here are three examples where problems can occur with each horizon. The future horizon is perhaps the most troublesome since by its nature an analysis is an educated guess. Where an analysis can run into trouble is when one price forecast is used to drive budget and another forecast is used to value a power contract. There should be consistency between the underlying assumptions—unless of course it was intended. The past can be equally troublesome. If someone wants to do an analysis on settlements data they need to know if the data most current (the market will send rebills). A more specific analysis that hangs up many inexperienced people is when looking at energy revenues. An inexperienced analyst would overlook make whole payments for those times the unit was asked to run uneconomically (below their bid). In the present an analyst can be looking for real time operations data. It is not too uncommon to have feeds for similar data like power plant output. Typically there are at least three meter feeds: the plant's internal meters, the market/iso meters and the metering authority meters. Which is the relevant one for the analysis? Is the meter a net or gross value? If something happens to the meter where you are receiving poor values who do you call? There are many examples of improper use of information. There is a cost to using the inappropriate data point.

Knowledge management is crucial for business continuity. How do we train someone new to the business? Is that an long or short process? How do we communicate process with each other. If something happens to a key individual how do we replace that individual? Knowledge is built up over time. There was a quote in a forensic engineering textbook I had read: "Good judgement comes from experience. Experience comes from poor judgment." How do we make sure good judgement is passed from person to person? How do we share the lessons learned? During the space race in the 1960s we had the technical knowledge to put a man on the moon. That knowledge is all but gone.

Strategies for managing information:

Create an information owner. Assign someone to be the go to person for settlements data, GADS data, forward curves, forecasts etc. They are responsible to provide the most up to date information and sign off on the data's use.

Incorporate the owner into the analysis. The owner should make sure the analysis is using the most appropriate data and that data is being interpreted properly.

Create datawarehouses. This does not have to be a formal IT solution. It can be excel files in a directory on the network or a more robust database. The key issue is to create central location where the information is stored and easily accessed. This will help facilitate quicker analyses and if designed well can reduce errors in utilizing the wrong data.

Strategies for managing Knowledge:

Cross train. Even without a formal process of recording knowledge. It is very important to cross train people inside and outside of their group function. Example group functions are structuring, strategy, commercial analysis, asset management, settlements or corporate development. This way an analyst in one group has a better understand of the information owned or used by another. Cross training adds redundancy in the event of a human resources issue.

Coordinate the information managers and users. This may sound bureaucratic but keeping your analysts and information managers informed of what other analyses are occurring elsewhere in the company is one way to make sure knowledge is shared and information is properly used. Someone in one group may be solving a problem that was similar to what is being solved somewhere else. The information managers can make sure that assumptions are properly characterized across all groups.

Record Knowledge. Very simply: write it down. Write down your process. Give detailed directions. There are some technical solutions that make this easy and accessible. One is a wiki (http://en.wikipedia.org/wiki/Wiki). Wikis are web based and extremely easy to use. They have good change management (records who makes the change, what they changed and when they changed it). We use wikis internally at Pharos to provide detailed descriptions of the content we are developing. It has become a source of internal training.