Next I my BAM Series, a post about BAM KPIs.
First the theory from the oracle documentation
There are two major types of performance indicators: outcome metrics and driver metrics. Outcome metrics, sometimes called lagging indicators, measure business activity that a strategy is designed to achieve, such as return rate per month or average call processing time for the last moving hour. Driver metrics, sometimes called leading indicators, measure business activity that influences the outcome metrics.
KPIs are outcome metrics that compare performance measures to specific business goals that roll up into larger organizational strategies requiring monitoring, improvement, and evaluation. KPI values vary with time, have targets to determine performance status, include dimensions to allow for more specific analysis, and can be compared over time for trending purposes and to identify performance patterns.
You can define two types of KPIs in BAM: Realtime KPIs and Scheduled KPIs. A Realtime KPI collects data and performs incremental calculations of a measure continuously. For example, a Judicial System might update the number of waiting asylum applications every time a request arrives. A Scheduled KPI collects data and recalculates the aggregation from a specific time at a regular interval. For example, a Judicial System might update the number of closed asylum applications every week.
A KPI’s status is determined by comparing its value against high and low thresholds that you define.
Risk indicators are driver metrics that provide early warning signs to identify events likely to jeopardize KPIs. For example, a risk indicator for a Judicial System might be the number of asylum applications in stock, which can block the completion of an application if it is too high. You can associate a risk indicator with a Realtime KPI.
In my post about queries (BAM Queries) I created a Group SQL Query (Open And Closed Tasks Without Timing) with no dimensions. I will use this to create a scheduled KPI. I also want to associate a risk indicator with a Realtime KPI, for this I create a Continuous Query (Open Task Monitoring).
After creating these KPIs I created a special view called a KPI Watchlist (Watchlist Tasks) that uses a group of related KPIs. Details about the KPI watchlist can be found in my BAM Views post. In my post about BAM Alerts I will also use the values of a KPI measure (high, medium, low) to initiate different alert actions.
Use of KPIs was not part of the POC that I’ve done, but I have added this for completeness as subject to this blog series.
I will now continue with some notes about the used KPIs.
KPI ‘AVG Time Open’
The KPI ‘AVG Time Open’ gives the average time usertasks are open. The threshold has a value of 10. This makes the target currently is 65%. As show in the following picture, an email with a dashboard link is send when the deviation from threshold is high (60% or more). Used for requirement 11 and 12 (BAM Projects requirements).
Note: To get this working emailing has to be right configured in your environment (not part of this post).
I have also attached the ‘OpenTaskMonitoring’ risk indicator. ‘OpenTaskMonitoring’ is a continuous SQL query (BAM Queries). If the number of open tasks is growing rapidly, this is a risk for the average running time of the tasks. This is shown in the Risks column (red = increased risk, green = okay). I’m not 100% sure if this is completely correct. Personally, I find this indicator is difficult to interpret. And the documentation is not very clear about this.
KPI ‘Number Of Open Tasks’
Next the KPI ‘Number Of Open Tasks’. This KPI gives the number of open tasks. Used for requirement 11.
In this blog I have briefly shown something of KPIs. Because KPIs were not part of the POC, I did not come into much details. Particularly around risks things are still not entirely clear.
In the next post I will do the Business views.
You can download the KPIs and the other related objects here.