This　paper　investigates　some　cooperative　Unmanned　Aerial　Vehicles　(UAVs)　executing　both　visiting　tasks　and　transportation　tasks.　The　UAVs　are　subject　to　random　shocks　which　may　harm　both　the　cargoes　on　the　UAVs　and　the　UAVs　themselves.　In　case　an　UAV　is　destructed,　subsequently,　all　the　cargoes　it　carries　will　be　damaged.　Furthermore,　if　only　some　cargoes　on　an　UAV　are　damaged　but　the　UAV　itself　is　still　functional,　then　the　UAV　can　still　continue　its　visiting　tasks.　In　order　to　reduce　the　probability　of　UAV　destruction,　an　UAV　is　allowed　to　abort　its　task　if　it　is　found　to　have　suffered　too　many　shocks　after　finishing　the　visit　of　a　certain　number　of　targets.　For　each　given　strategy　of　UAV　routing　and　loading,　the　optimal　aborting　strategy　needs　to　take　into　consideration　both　the　expected　cost　due　to　UAV　destruction　and　the　expected　cost　due　to　unfinished　tasks.　Eventually,　the　optimal　UAV　routing　and　loading　strategy　can　be　solved　assuming　that　the　optimal　aborting　strategy　is　always　adopted,　with　the　objective　to　minimize　the　total　cost　of　UAVs　destruction,　total　cost　of　damaged　cargoes,　and　the　expected　cost　of　unvisited　targets.　Monte　Carlo　simulation　is　applied　to　evaluate　the　cost　of　damaged　cargoes,　and　tabu　search　algorithm　is　employed　to　optimize　the　routing　of　UAVs.　Case　studies　are　shown　to　illustrate　the　application　of　the　framework.　©　2020　Elsevier　Ltd