This　paper　investigates　an　extended　bike-sharing　rebalancing　problem　called　the　bike-sharing　rebalancing　problem　with　variable　demand　(BRPVD),　which　considers　the　influence　of　the　number　of　bikes　distributed　by　the　operator　on　users’　demand.　The　solutions　to　the　BRPVD　aim　at　maximizing　the　profit　of　bike-sharing　operators　by　making　routing　plans　of　transport　vehicles　and　determining　the　target　number　of　bikes　at　each　station　after　redistribution　operation.　The　problem　is　modeled　as　a　mixed-integer　non-linear　programming　model　and　is　linearized　into　a　mixed-inter　linear　programming　model.　The　model　is　tested　in　cases　adopting　real　Beijing　Mobike　data,　revealing　that　the　rebalancing　behavior　significantly　impacts　users’　demand,　hence　guiding　the　operator.　©　2021