Understanding Business Intelligence Ecosystem for Corrections

For the corrections agency seeking to maximize the value of its OMS, business intelligence and data analytics solutions are moving from ‘nice to have’ to essential components of the information strategy.

Given the breadth and complexity of business problems facing corrections leaders, and the demand for actionable, fact-based information from across the full range of information stakeholders, business intelligence (BI) and data analytics tools are increasingly viewed as core components of the agency’s information infrastructure. Indeed the majority of corrections professionals I meet at conferences and other gatherings have a basic understanding of the potential that such tools represent in their efforts to drive greater insights into their data.

Yet, it is just as common that they lack a full understanding of the full range of capabilities and uses for these powerful tools. Often, I am asked such questions as ‘Is it a database?’ Is it a statistical analysis solution? Is it an operational data reporting tool? Is it an information dashboard? Is it a decision support system? Is it a forecasting tool? Is it a trend analysis function? Is it a predictive analysis solution? The answer to each of these questions is ‘yes,’ as all the leading tools include these capabilities. However, the secret to their successful application in the corrections environment requires that they be part of a well-defined, structured BI ecosystem that is closely aligned with the OMS and the agency’s overall information strategy.

For some agencies, the first significant investment in BI and data analytics comes as part of the OMS modernization process. For others, the investment is undertaken to extract greater value from the existing OMS assets. Either way, for the CIO or senior analytics manager in the agency responsible for planning and execution, it is vitally important to understand the full spectrum of the BI ecosystem in the context of the agency’s needs prior to its selection and implementation. A successful BI solution, effectively implemented and strategically managed, will help the agency to achieve its key goals, such as improvement in operational efficiency; officer and offender safety, as well as community safety; offender programs and rehabilitation; recidivism reduction strategies; and threat minimization or mitigation, such as early identification of potential gang recruitment or offender radicalization. All of these can be effectively addressed through an effective BI and data analytics strategy and ecosystem.

A typical BI ecosystem is composed of four main areas, which are classified based on the nature of its functions, namely:

• Descriptive – Insight into the past, WHAT has happened?
• Diagnostic – Analyze the past, WHY has it happened?
• Predictive – Understand the future, WHAT could happen?
 Prescriptive – Advise on possible outcomes, WHAT should we do?

Descriptive Analytics: Insight into the past

The most common, basic and typical usage of BI systems is descriptive analytics. As its name implies, descriptive analytics essentially “describes” the situation, either in a current context or in the past or both, based on the data associated with the actual events that occurred. Descriptive analytics summarizes the raw data, connects the dots and transforms it into actionable information. An example might be an analysis of the number of male offenders from a particular facility that have recidivated over a rolling 24-month period. In descriptive analytics, the raw operational or transactional data, which are highly technical in nature, are transformed into information, which can be easily consumed, interpreted and, if appropriate, acted upon by the user.

Descriptive analytics provides insights into historical data. The past can refer to any point in time at which an event has occurred, whether it is one minute ago, or one year ago. Descriptive analytics are useful because they allow us to learn from past activities and results, and understand how they might influence future outcomes.

In the corrections context, it is critical to have insights into data from facility operations to ensure the safety of staff and offenders, as well as the general public. Using descriptive analytics techniques, the agency can be more proactive in its efforts to achieve desired outcomes. For example, agency staff can gain a broad view of an offender, such as the number of incidents, the number and frequency of grievances raised, jobs assigned and completed, and program attendance, to better plan for the offender’s eventual re-entry into the community.

Descriptive statistics are useful to understand key metrics as well, such as facility demographics, number of incidents logged by location, and number of grievances raised per shift.


Typically, the descriptive analytics component of a BI and analytics strategy is enabled through implementation of enterprise data warehouse (EDW) and/or data marts to manage the data. All the associated information, findings and outcomes of descriptive analytics are presented and disseminated using reports, as well as information dashboards. Visual and interactive representations of the information are viewed as a more effective way of presenting the findings, when compared to traditional static, tabular reports.

Diagnostic Analytics: Analyze the past

Diagnostic analytics is an extension of descriptive analysis, where the user leverages the descriptive data (information) to dig into and analyze the root cause. Diagnostic analytics are useful for agencies as they attempt understand the reason(s) behind a particular event or pattern of events. For example, following an upward trend in the number of incidents at a given location, an analytic specialist can slice and dice the data across numerous factors to determine in a fact-based manner the likely causes for the rise. The analysis might reveal that the incident trend is related to staffing issues (e.g. insufficient training), lack of adequate medical services, or meal services that are consistently late. Such fact-based conclusions can be shared with appropriate agency staff who can take necessary actions to address and resolve the situation. A proactive agency might then use the information to establish key performance benchmarks and monitor future activity to determine if the initial issue has been fully resolved – in this instance demonstrating that logged incidents are both trending downward and are below tolerance thresholds.

Predictive Analytics: Understanding the future

Through application of scientific approaches and mathematical modelling, predictive analytics can essentially ‘predict’ possible outcomes based on the results of historical data mining and patterns. Predictive analytics can help agencies in their efforts to understand the future and take necessary proactive or risk mitigation measures. Predictive analytics provides agencies with actionable insights based on the true events (data) recorded and provides estimates about the likelihood of a future outcome. Though there is not 100% certainty that the predictions will in fact become a reality, a high degree of probability can be expected if the agency and the analyst leverage a quality data set, sound predictive models, algorithms and analytic rigor. As new, unanticipated events unfold in the future, they can be ‘learned’ and incorporated into the data set to strengthen the predictive models.

By combining descriptive data and predictive analytical models, agencies can predict the likely outcomes, such as the recidivism rate, and then look at the variables and factors that that contribute to it to understand if specific, corrective actions can be taken. Agencies can apply predictive modelling and analytics at any time to look into the future and make recommendations – in capacity planning, for example – that will contribute to the overall efficiency and effectiveness of the agency’s performance.

Prescriptive Analytics: Advise on possible outcomes

A new paradigm in the BI ecosystem or landscape is to enable the data analytics platform to ‘prescribe’ several different potential actions that could be considered and applied as a potential solution for a problem that is foreseen during predictive analysis. Prescriptive analytics quantifies the results of the solution(s) prescribed, which can then help the agency to measure the success of the solution and make a proper decision before it is applied.

For example, using predictive models agencies will be able to score the risk of recidivism that could occur in the future, and by applying prescriptive models, make specific recommendations or take specific actions that will minimize the risk from occurring. Such recommendations might pertain to a specific offender, such as completion of an anger management course three months prior to release, or could be shared with third-party non-profits that work with groups of offenders to ensure their success in the community after release.

As agency leaders are increasingly challenged to effectively address any number of structural and operational issues related to performance, the need for powerful tools to investigate and model data will become ever more important. Similarly, the skills that agencies will need to acquire to perform such advanced analysis will be a core component of a successful information strategy. In future posts on BI and analytics, I will write in detail about those skillsets and ways to potentially organize analytics teams, as well as more in-depth commentary on data governance, automation of data and analytics, and more.

I welcome your comments and feedback on this post. Please feel free to e-mail me at MGovindappa@abilis-solutions.com.

© 2017 Abilis Solutions. All rights reserved.

About the Author

Maruthi Govindappa

Maruthi leads the Business Intelligence and Data Analytics practice at Abilis, with a deep focus on developing strategies and planning implementations to address enterprise business intelligence needs for correctional and public safety agencies. Maruthi specializes in building advanced, custom E2E BI solutions, state of the art visual dashboards, utilizing in the process cutting-edge BI tools that are customized for specific customer needs.

Maruthi has more than 12 years of experience in the Business Intelligence domain, implementation of large BI solutions for various large-scale enterprise organizations. He has an engineering degree from VTU (Visvesvaraya Technological University) of India. Maruthi is very passionate about BI and the value it delivers to the customer. He is a member of The Data Warehouse Institute (TDWI) and a regular participant in various BI forums. He also led and continues to participate in technology user groups for Talend, Tableau and Alteryx.

Lire la suite

Laisser un commentaire

Votre adresse e-mail ne sera pas publiée. Les champs obligatoires sont indiqués avec *