18th century Scottish poet Robert Burns writes on the nature of plans in his poem To a Mouse. While apostrophizing a mouse’s ability to live in the present, Burns discusses his belief that schemes are made in between regretful memories and a fear of an uncertain future. He laments his plans as guesses.

However, Burns’s mentality about plans exemplifies our misunderstanding of what a plan actually is and why they ostensibly “go askew.” He writes:

But little Mouse, you are not alone,

In proving foresight may be vain:

The best laid schemes of mice and men

Go often askew,

And leave us nothing but grief and pain,

For promised joy!

Is that so, Robert Burns? The best laid schemes (plans) go often askew? Maybe that depends on how your planning method, in which there are two types.

Prescriptive-deductive planning is the right handed method of mapping the future — dextroplanning if you will. It infers events, facts, and patterns through a primary theory of reality as it is and reality as it will be. In essence, it’s planning by scientific method.

When people, like Robert Burns, believe a plan went wrong, they are mistaken. The plan is inherently neutral, but its inevitable confrontation with observable and unknowable information is interpreted by planners as bad. Prescriptive-deductive plans are eventually contradicted by overlooked and unforeseeable circumstances. In other words, saying a best laid plan went askew assumes the plan was ever based on reality-as-it-was and reality-as-it-would-be.

In times of uncertainty such as a world experiencing a pandemic, prescriptive-deductive planning should be used cautiously, because it approaches the future with an assumed understanding of what reality is and what reality will be before considering pertinent events, facts, and patterns. A solution for planning in times of uncertainty is to make an adaptive plan that undergoes continuous evaluation through continuous observation to form a theory. This approach is opposite to how most people plan for future events. It is descriptive-inductive.

Descriptive-Inductive planning is the left handed method of mapping the future — levoplanning. It infers theory of reality-as-it-is and reality-as-it-will-be through the primary observation of events, facts, and patterns.

Most of our exposure to descriptive-inductive planning is through weather forecasting. While meteorologists use prescriptive-deductive models to forecast probable weather conditions from hours to months away, their forecasts are constantly being updated with “current conditions,” which are data sets that describe rapidly growing information to induct observable patterns into generalized conclusions.

Emerging applications of descriptive-inductive planning is in a concept called “nowcasting,” in which epidemic conditions are monitored using social media and reputable databases. Epidemiologists use prescriptive-deductive models that forecast flu outbreaks, but it is through descriptive-inductive models that they nowcast flu outbreaks as they are geographically occurring. This is yet to be a reliable science, but it has potential in the growing field of artificial intelligence, which unfortunately is also its current drawback.

Descriptive-inductive planning requires labor and machine power — tons. In addition, when observations of events, facts, and patterns are flawed, we can mistakenly see a story where there isn’t one. In fairness, however, an equal tradeoff in prescriptive-deductive planning is the tendency to empirically observe sets of events, facts, and patterns that validate a preconceived story while ignoring information that is contrary to theoretical assumptions.

A balance can be found however. During times of certainty, use the right handed planning method: prescriptive-deductive. During times of uncertainty, use the left handed planning method: descriptive-inductive. Sometimes reality is stable, and sometimes it isn’t. Crossfading between these two planning methods will asymptotically correct forecast rates of error into optimization. Using one and not the other is regressive optimization, in which a fall backward is required to gain an optimizable step forward. Using both in conjunction provides progressive optimization in which steps forward will diminish in progressive returns at an environmentally-caused rate until we reform the plan into greater leaps in absolute value — the process is cyclical. We are exponentially departing from the dark while we are asymptotically approaching the light. The difference between the asymptote and the function is our infinitely and fractally belief about the world and the world itself. Always closer, forever distant.