This article addresses the one-commodity pickup and delivery traveling salesman problem (1-PDTSP), which is a generalization of the well-known traveling salesman problem. The 1-PDTSP aims to find a Hamiltonian tour in which a set of supply points (pickup locations), demand points (delivery locations) are visited and, the total traveled distance is minimized. We propose a hybrid metaheuristic based on multi-start evolutionary local search and variable neighborhood descent to solve the 1-PDTSP. To test the performance of our algorithm, we solve instances with up to 500 nodes available in the literature and we demonstrate that our approach is able to provide competitive results when comparing to other existing strategies. Since a direct application of the 1-PDTSP arises as the bicycle repositioning problem, we also use our metaheuristic algorithm to solve a set of real-case instances based on EnCicla, the bicycle sharing system in the Aburra Valley (Antioquia, Colombia).

中文翻译：

---

单品提货旅行商问题的多起点进化局部搜索

本文解决了单一商品取货和送货旅行商问题 (1-PDTSP)，它是著名旅行商问题的推广。1-PDTSP 旨在找到一个哈密顿旅行，其中访问了一组供应点（取货地点）、需求点（交货地点），并且总行驶距离最小化。我们提出了一种基于多开始进化局部搜索和可变邻域下降的混合元启发式算法来解决 1-PDTSP。为了测试我们算法的性能，我们解决了文献中多达 500 个可用节点的实例，并证明了与其他现有策略相比，我们的方法能够提供有竞争力的结果。由于 1-PDTSP 的直接应用作为自行车重新定位问题出现，