Many real-world QAP applications (e.g., arranging computer components on a circuit board to minimize wiring, or locating hospital departments to minimize patient travel) involve physical distances that are metric and often approximately linear along a corridor. Kelk’s bound gives practitioners a baseline: if a sophisticated metaheuristic (like a genetic algorithm) produces a solution more than twice the LAP lower bound, it is performing worse than a simple, provable reduction.
The reason Kelk 2007 stayed relevant for so long is its unique engine, which allows for customization that standard OpenType fonts simply cannot match: kelk 2007