Originated by Genrich Altshuller (Algorithm of Invention, 1969).
A list of forty recurring solution patterns, such as segmentation, asymmetry, and prior action, drawn from patents and used as prompts to resolve a design contradiction.
One contradiction, a box that must be stronger and lighter, and four principles to resolve it
A meal-kit company wants a returnable shipping box that survives twenty trips, and every stronger version so far has been heavier, which raises the postage. Pick one of four principles from the forty: the line under it gives the box that principle produces and says whether it removes the contradiction or only trades one cost for another. Illustrative.
Apply the principle
- Pick a principle to see the box it produces.Divide the object into parts. A thin sleeve with a clip-in rigid corner frame: only the frame carries the stacking load, and a worn frame is swapped alone. Trades it off: 20 percent lighter at the same strength, but packers lose ten seconds a box.Make each part fit its own job. Heavy board only on the base and corners, where drops and stacking hit, light board on the walls and lid. Removes it: the same drop rating at 30 percent less weight, on the existing machines.Put one object inside another. Tapered walls so empty boxes stack inside each other for the trip back. Moves it: each box weighs the same, but ten empties ship in the space of three, so only the return leg gets cheaper.Replace a uniform material with a composite. Corrugated plastic on a honeycomb core. Removes it at a price: forty trips at the old weight, but three times the unit cost, so it pays only if returns run above 90 percent.
- Divide the object into parts. A thin sleeve with a clip-in rigid corner frame: only the frame carries the stacking load, and a worn frame is swapped alone. Trades it off: 20 percent lighter at the same strength, but packers lose ten seconds a box. (p: 1 Segmentation)
- Make each part fit its own job. Heavy board only on the base and corners, where drops and stacking hit, light board on the walls and lid. Removes it: the same drop rating at 30 percent less weight, on the existing machines. (p: 3 Local quality)
- Put one object inside another. Tapered walls so empty boxes stack inside each other for the trip back. Moves it: each box weighs the same, but ten empties ship in the space of three, so only the return leg gets cheaper. (p: 7 Nesting)
- Replace a uniform material with a composite. Corrugated plastic on a honeycomb core. Removes it at a price: forty trips at the old weight, but three times the unit cost, so it pays only if returns run above 90 percent. (p: 40 Composites)
An example
A meal-kit company wanted a returnable box that could make twenty round trips. Its packaging engineer had spent a quarter on the obvious path: thicker board, then double walls, then a plastic liner. Each version was stronger and heavier, and postage was charged by weight, so every gain in trips was eaten by shipping cost.
She wrote the problem as a contradiction: improve strength without worsening weight. Then she picked four of the forty principles and forced each onto the box. Segmentation gave a thin sleeve with a clip-in corner frame. Local quality gave heavy board only on the base and corners, light board elsewhere. Nesting gave tapered walls so empties stacked inside each other for the trip back. Composite materials gave corrugated plastic on a honeycomb core. She compared the four sketches against the twenty-trip target and the postage budget. Local quality won on cost and kept the existing supplier, and nesting was added to cut the return leg; together they hit the target at a lighter weight than the original box.
Why it works
Most design work treats a contradiction as a dial: more strength costs more weight, and the job is finding the least bad setting. The principles start from a different observation. The same few moves have resolved the same kinds of conflict across thousands of unrelated inventions, so a conflict that looks unique to one product has usually been beaten elsewhere, by a pattern that can be named.
The list works by forcing. Applying a principle such as segmentation to a specific box produces an idea the engineer would not have reached by turning the dial, because the dial only offers more or less of the current design. Several principles applied in turn yield several concrete sketches, and some of them escape the tradeoff instead of splitting the difference.
The thinking behind it
Genrich Altshuller Genrich Altshuller, a Soviet engineer and inventor, built TRIZ, the theory of inventive problem solving, from the study of thousands of patents. He held that technical systems evolve by knowable patterns and that inventive problems turn on contradictions, which a small set of principles can resolve. He set the list of forty inventive principles in 1971, after Algorithm of Invention (1969), and presented the wider theory in Creativity as an Exact Science (1979). Beside the principles, TRIZ offers separation of conflicting requirements and ARIZ, his step-by-step algorithm for harder problems.
How to run it
- State the contradiction. Write what must improve and what gets worse when it does, in the terms of the product, such as strength against weight.
- Find the suggested principles. Look up the pair in a contradiction table, or scan the full list of forty when the pair is unclear.
- Force each principle onto the case. Ask what the principle would mean for this exact object, even when the first answer seems absurd.
- Sketch a concrete design for each. Write enough detail to estimate the gain and the cost, such as weight, trips, and unit price.
- Check whether it escapes the tradeoff. Separate designs that remove the conflict from those that only move it somewhere else.
- Compare and combine. Pick the most promising sketch against the original targets, and look for a second principle that adds to it.
Where it goes wrong
The common failure is a fuzzy contradiction. Stated as make it better, the problem matches every principle and none of them helps. Write the conflict as two named properties, one improving and one worsening, before opening the list.
The second is reading the principles as answers. Each is a prompt, and segmentation means nothing until someone works out what it would do to this box. A session that ends with a list of principle names and no sketches has skipped the work.
The third is stopping at the first principle that seems to fit. The value comes from several sketches compared side by side; the obvious principle often only moves the tradeoff. Apply at least three before choosing.
When to use something else
- Separation Principles. When the conflict is one property that must have two opposite values, which can be separated in time, space, or scale.
- SCAMPER. When the aim is a quick spread of variations on an existing product without a defined technical conflict.
- TRIZ. When the problem needs the whole method, from the ideal result and resources to the evolution of the system.
Where it sits in SOLVE eX
Go to the source
- Algorithm of Invention, 1969