Probability Management is rethinking uncertainty through communication, calculation, and credible estimates.
How many boxes of Girl Scout cookies should a troop purchase for the cookie drive in the face of uncertain demand and available volunteers?
Where should an energy firm explore for oil and gas in the face of uncertain geology, prices and environmental concerns?
Which investments should we choose for our retirement funds in the face of uncertain markets?
Such uncertainties drive decision making at all levels of business, government and personal life, where they are often replaced with single "average" best guesses. Unfortunately this leads to a set of systematic mathematical errors known collectively as the Flaw of Averages, which explain why so many things are behind schedule, beyond budget and below projection.
What is a SIP?
A SIP, or Stochastic Information Packet, is a way of representing an uncertainty as a data array comprised of thousands of possible outcomes. A SIP makes the abstract concept of a probability distribution actionable, additive, and auditable as discussed below.
Communication of Uncertainties
For example, the uncertainty of rolling a die could be communicated as an array of thousands of actual or simulated die rolls, which could be stored in Excel or a database. The open SIPmath™ Standard enables legacy and future simulation models to communicate with each other, ushering in a new paradigm for enterprise risk management.
Calculation with Uncertainties
Calculating with SIPs is called SIPmath, and it lets you perform arithmetic on uncertainties in your native spreadsheet just as you would with numbers. Thus SIPs are actionable. This leads to better decisions in investment, purchasing, and scheduling, etc. In a recent breakthrough, Microsoft Excel has become robust enough to interactively process SIPs of thousands of trials, placing the benefits of SIPmath within reach of tens of millions of managers, scientists, engineers and educators. SIPmath has even been used to teach investment theory to middle school students. The open SIPmath standard is not limited to spreadsheets, but is easy to implement in any software environment that supports arrays. This allows the output SIPs of individual risk models to be aggregated into enterprise wide models. Therefore, they are additive.
Credibility of Uncertain Estimates
Because uncertainties are stored as unambiguous data with provenance, statistical experts may publish their results for use within risk dashboards by non-experts. For example, the Federal Reserve models the uncertainty of future GDP, inflation, etc. The US Geological Survey models uncertain earthquake magnitude for any latitude and longitude in the US. Financial data firms offer details on the uncertainty of future asset values. Trusted sources such as these give management the "permission" to be uncertain within auditable limits.
It's Nothing New
Computers, mobile phones, and touch screens existed before the smart phone, so what’s the big deal? The big deal is that the smart phone became a node in a network of 100 million other smart phones that allowed people for the first time to share and process photos, maps, traffic conditions, and many other applications.
Monte Carlo simulation, array arithmetic, and databases existed before probability management, so what’s the big deal? The big deal is that by creating open cross platform standards, every computer becomes a node in a network of other computers that allow people for the first time to unambiguously share uncertainty and mitigate the Flaw of Averages at all levels.