Deductive reasoning (sometimes referred to as “deduction”) occurs when two or more previously proven assertions are used to draw a demonstrably related conclusion1. In most cases, deduction is what we think of when we talk about Engineering processes that arrive at valid conclusions (e.g., sizing, derivation, and verification).
In classical logic2 each assertion is usually referred to as a “premesis”3, meaning that they have previously been subjected to individual validation.
See also reasoning. Contrast with abductive reasoning and inductive reasoning.
Deductive logic, when thoroughly expressed, is a form of model built on established facts and data without the use of assumptions. “Established facts and data” include, for example, “first principles”, formally qualified data, and (in some circumstances) guaranteed claims of performance by a commercial supplier.
In theory, deduction is the preferred type of reasoning involved in Engineering because it inherently results in certitude of the conclusion. As a practical matter, however, inductive logic is commonly relied on to deal with certain “off nominals”. In particular, we rarely examine all combinations of tolerance, circumstance, and uncertainty when qualifying a design; we usually “assume” that they can’t all go to the adverse extremes at one time, which is an inherently inductive thought process.
Footnotes