QHL: A Fast Algorithm for Exact Constrained Shortest Path Search on Road Networks

Abstract

Route planning is fundamental in our daily life. However, existing mapping applications focus on recommending routes by optimizing one single objective, which is inconsistent with some scenarios where users prefer the optimal route under a constraint. The constrained shortest path (CSP) query matches this requirement, but the query efficiencies of previous solutions are often low due to CSP's NP-hardness. In the era of big data, state-of-the-art indexes are getting larger to support faster query processing. Recent attempts to preprocess more intermediate results and reduce the number of table lookups have proved successful in solving the CSP. However, the best-known algorithm ignores some information in the CSP queries and tries to solve a more general problem before tackling the exact CSP. In this paper, we propose by far the fastest algorithm called QHL, which fully utilizes the pruning power of the CSP query information. Specifically, we preprocess our index by generating pruning conditions that can improve query efficiency. We also conducted extensive experiments on real-world datasets to demonstrate the superiority of our proposed algorithm. QHL could answer each CSP query in around 50 μs and run faster than the best-known algorithm by orders of magnitude.