Within 150 years of Benjamin Franklin’s reckless kite experiment, which proved “The Sameness of the Electric Matter with that of Lightning,” electrification was distributing electric matter directly to homes and factories for use in lightbulbs and machinery. Now, in a process we call Chancification, three complementary open technologies are creating the power grid for probability, allowing managers without statistical training to calculate the chances of avoiding risks and achieving their goals.  

Three battle hardened veterans of the War on Averages with over 90 years of consulting experience between them have teamed up to create information systems for uncertainty. The common enemy is the Flaw of Averages, a set of systematic errors that occur when uncertainties are represented by single “average” numbers. The new approach defines data structures that obey both the laws of arithmetic and the laws of probability, unleashing the potential of Decision Analysis, Portfolio Theory, Options, Information Economics and other resilient techniques for managing opportunities and risks in an increasingly uncertain world.

The Team

Sam Savage, Executive Director of 501(c)(3) nonprofit ProbabilityManagement.org (Wikipedia), author of The Flaw of Averages, Adjunct Professor of Civil & Environmental Engineering at Stanford University, and inventor of the Stochastic Information Packet (SIP).

Doug Hubbard (Wikipedia), Founder of Hubbard Decision Research, author of How to Measure Anything (with over 100,000 copies sold) and The Failure of Risk Management, and inventor of Applied Information Economics and the HDR pseudo-random number generator.

Tom Keelin, former Worldwide Managing – Director of the Strategic Decisions Group, Managing Partner of Keelin Reeds Partners,  and inventor of the Metalog Distribution (Wikipedia).

Potential Applications

Chancification takes computer simulation from siloed applications to collaborative networks. Potential applications include:

• Linking national weather simulations to power grid simulations to estimate the chance of collapse due to failed equipment, excess heating or air conditioning load.
• Aggregating risks across infrastructure networks to mitigate the chance of safety risks and minimal cost.
• Using crowdsourced data on forecasting errors to estimate the chance of achieving projected tax revenues.
• Linking the results of ensembles of COVID 19 models at the CDC to local models to predict the chance of exceeding ICU capacity.
• Using military aircraft reliability data to estimate the chance of being ready fly a mission at a moment’s notice.

Connecting the Levels of the Organization

Chancification impacts multiple levels of an organization.

1. Statisticians, data scientists and other analysts will now be able to deliver their results as SIP Libraries to drive chance-based decision dashboards.
2. Mid-level managers will use the SIP Libraries generated by analysts much as the general public uses electricity generated in power stations to illuminate the chances of achieving targets and avoiding pitfalls.
3. Top level managers, even those with Post Traumatic statistics disorder (PTSD) will be able to address front line decisions in terms of both quantified risk and return, just as Wall Street investors have, since the advent of Modern Portfolio Theory in the 1960s.

 

Teaching the ABCs of Chancification

The steps of Chancification are:

1. Assess the salient uncertainties
   Managers accustomed to requesting and receiving numbers must be trained to detect the most important types of uncertainties.
2. Bound the uncertainties
   Bounds on the uncertainties must be quantified. Many analysts are doing this already, but in most cases their work can be greatly simplified with the revolutionary Metalog distribution.
3. Correlate the variables
   It is critical to account for the interrelationships between uncertainties, managers rooted in the world of numbers are often oblivious to this concept. Analysts are aware of the concept, but need to be taught how Chancification handles this in a new robust manner
4. Deliver stochastic libraries
   Analytical packages such as Python and R, as well as Excel, can produce SIP libraries for use in the cloud or behind firewalls.
5. Employ the results to improve decision making
   ChanceCalc can link to the libraries delivered above, and then estimate chances of achieving targets or avoiding pitfalls with a single command.

We can introduce the appropriate aspects of Chancification to all levels of management and point your organization toward the open technologies upon which it is based.

Find out more https://www.probabilitymanagement.org/precise-uncertainty