As a data scientist, I recently explored the economic challenges of the USSR through the book Red Plenty, which details how centrally planned resource allocation faced significant obstacles. The Soviet system relied on manual balance sheets to allocate resources across hundreds of thousands of commodities, leading to inefficiencies.

Administrators simplified calculations by only tracking the 10,000 most important commodities, causing hidden shortages of untracked inputs. They also aggregated diverse products like steel tubes by tonnage, resulting in surpluses of thick tubes and shortages of thin ones. Adjustments were only propagated a few supply chain levels deep, creating cascading shortages and idle factories.

In the late 1960s, economists known as the "optimal planners" introduced linear programming, pioneered by Leonid Kantorovich, who won the 1975 Nobel Prize in Economics. This mathematical approach allowed optimization of resource allocation across multiple factories and products. In the steel industry, it helped distribute 10,000 product types among 500 producers based on efficiency.

Despite these advances, implementation remained limited due to political resistance, preventing many model recommendations from being enacted.