To appreciate Kelk’s contribution, one must first understand the QAP's unique difficulty. Unlike the Linear Assignment Problem (which can be solved in polynomial time), the QAP is not only NP-hard but also NP-hard to approximate within any constant factor (unless P=NP). This means there is no polynomial-time algorithm that guarantees a solution within, say, 1000 times the optimal value for all instances. This stark inapproximability forces researchers to either focus on special cases (e.g., when the flow or distance matrices have specific properties) or to seek approximation algorithms with guarantees that depend on instance parameters.
The Linear Arrangement Problem (LAP) , a simpler relative, asks: given a graph, order its vertices linearly to minimize the sum of the absolute differences of indices for each edge. The LAP is also NP-hard but has known constant-factor approximation algorithms (e.g., the 2-approximation by Rao and Richa). A natural question arises: can we solve a QAP instance by reducing it to an LAP instance? Kelk (2007) provides a definitive, nuanced answer to this question. kelk 2007
Over the years, several myths have emerged regarding "Kelk 2007." Let's debunk them: Into this fray stepped Kelk
To appreciate "Kelk 2007," we must rewind to the mid-2000s. Computational power was increasing exponentially, but algorithms were struggling to keep pace with complex physical problems. Specifically, researchers faced three major challenges: which was computationally expensive)
Into this fray stepped Kelk. Unlike previous works that focused on monolithic solvers (solving fluid and structure simultaneously, which was computationally expensive), Kelk proposed novel partitioned iterative schemes. His 2007 thesis provided rigorous proofs for stability conditions that had previously been observed only empirically.
Given its academic value, "Kelk 2007" is not behind a paywall in the same way commercial journals are. You can typically access the full text via:
Note: As a private equity-backed entity, full public filings are limited. The following figures are based on investor reports and audited summary statements.
No products in the cart.