For　better　solving　vehicle　routing　problem　(VRP),　the　solution　processes　of　three　algorithms　including　ant　colony,　particle　swarm　optimization　and　simulated　annealing　are　introduced　and　compared　in　performance.　Three　customer　sizes　of　Solomon　data　set　are　selected　to　evaluate　the　solution　performance　of　these　algorithms.　Results　show　that　particle　swarm　optimization　algorithm　has　unsatisfactory　effects　in　capacitated　vehicle　routing　problem　(CVRP)　of　all　scales.　Simulated　annealing　algorithm　could　obtain　the　optimal　solution　in　small　and　medium-size.　Ant　colony　algorithm　has　the　highest　comprehensive　evaluation　for　solving　all　sizes　CVRP.　The　research　results　could　provide　reference　value　for　selecting　algorithms　to　solve　VRP　problem　with　volume　constraints.　©　2023,　The　Author(s),　under　exclusive　license　to　Springer　Nature　Singapore　Pte　Ltd.