KDnuggets Home » News » 2015 » Dec » Opinions, Interviews, Reports » Kanri Distance approach for translating Predictive Models to Actions ( 15:n42 )

Kanri Distance approach for translating Predictive Models to Actions


Kanri proprietary combination of patented statistical and process methods provides a uniquely powerful and insightful ability to evaluate large data sets with multiple variables.



By Rajesh Jugulum and Don Gray, Kanri Insights.

Kanri’s proprietary combination of patented statistical and process methods provides a uniquely powerful and insightful ability to evaluate large data sets with multiple variables.  While many tools evaluate patterns and dynamics for large data, only the Kanri Distance Calculator allows users to understand where they stand with respect to a desired target state and the specific contribution of each variable toward the overall distance from the target state.  The Kanri model not only calculates the relationship of variables within the overall data set, but more importantly mathematically teases out the interaction between each of them.

This combination of relational insights fuels Kanri’s breakthrough distance calculator.  It answers the question “In a world of exponentially expanding data how do I find the variables that will solve my problem?” and it helps quickly to reach that conclusion.  But the Kanri model does not stop there.  Kanri tells you exactly, formulaically how much each variable contributes.

The Kanri Distance Calculator opens a new world of solution development possibilities that can apply the power of massive data sets to an individual…or to an individualized objective.

KDC Approach

The purpose of Kanri business model is to provide a management system based on several attributes (variables) that are related.  Based on the information on these attributes a multivariate measurement scale is developed to measure the distances from a target or reference group.

After developing and validating the measurement scale, a root cause analysis (RCA) is performed on individual basis.  In RCA, we calculate contribution ratios (CRs) of the attributes.  This will help us to focus on those attributes that have highest impact through statistical process control (SPC) charts.

The detailed steps in the Kanri approach are given below:

  1. Define the target (healthy) group and all health attributes for consideration.
  2. Development of multivariate measurement scale with target group as the reference.
  3. Optimize the number of variables that are required for the measurement system
  4. Calculate each participant’s distance from target by running data through KDC.
  5. Perform individual diagnostic and identify contributing variables for each participant.

Why KDC?

  • Unique, fast, simple and flexible root cause identification system through a “distance measure” and methodology.
  • Allows to define success at individual levels and to continuously evolve the goals.
  • Provides quantitative and visual management system.
  • Helps build out extremely robust management profile for use with the professional support team.
  • Gives a measurement scale with a “distance measure” that is obtained based on multiple attributes.
  • Helps in understanding correlations between attributes (both within and between individuals).

Advantages of KDC

  • KDC produces distance to target scores for multivariate problems, even for problems that have been previously considered too difficult to solve.
  • KDC contemplates and quantifies the interactions or correlations between each of the variables.
  • KDC quickly identifies key variables regardless of the amount of variables contemplated.  Within days, instead of weeks or months, data sets can be narrowed not just at a population level but also for each individual participant.
  • KDC provides simple, easily understood outputs.  Outputs from other techniques can be very difficult to understand and interpret, especially between many variables.
  • KDC delivers quantitative data analysis for complex multivariate studies.  Most of the multivariate methods classify whether a person belongs to healthy vs. unhealthy group, but does not provide individualized data analytics for each participant.
  • KDC is distribution-type agnostic.  KDC is a non-probabilistic diagnostic engine.
  • KDC tests multiple simultaneous hypotheses.
  • KDC works well with both continuous and non-continuous data.