With　the　development　of　the　manufacturing　industry　and　the　gradual　improvement　of　highway　infrastructure　in　our　country,　many　manufacturing　firms　have　to　expand　their　markets,　but　no　longer　limited　to　the　local　market.　With　the　expansion　of　the　market,　the　difficulty　of　vehicle　management　of　enterprises　has　risen　sharply.　One　of　the　vehicle　management　problems　faced　by　enterprises　is　the　allocation　of　drivers.　In　actual　production,　the　allocation　of　order　drivers　is　often　done　by　manual　allocation,　which　will　cause　unreasonable　allocation　results　due　to　some　human　subjective　factors,　thus　causing　losses　to　enterprises.　Firms　want　to　allocate　orders　and　improve　customer　satisfaction　by　selecting　the　best　drivers.　For　better　recommendation,　this　paper　uses　machine　learning　and　other　technologies　to　study　and　design　the　driver　evaluation　formula,　and　realizes　the　driver　recommendation　function,　and　carries　out　simulation　experiments　to　simulate　and　verify　the　algorithm　used.　Through　the　accurate　recommendation　of　drivers,　the　distribution　quality　and　efficiency　of　enterprises　are　higher,　the　probability　of　accidents　and　violations　in　the　distribution　process　is　reduced,　the　operating　cost　of　enterprises　is　reduced,　and　more　benefits　are　brought　for　enterprises.　©　2023　IEEE.