The　assessment　of　potential　and　utilization　of　solar　energy　for　each　building　has　become　an　essential　precondition　of　urban　sustainable　development.　Currently,　there　are　two　challenges,　namely　high　computation　cost　for　an　assessment　within　a　large　city　and　missing　rooftop　information　for　each　building.　In　this　study,　by　taking　the　downtown　area　in　Shanghai　as　sample,　a　grid　scheme　with　buffer　was　proposed　to　alleviate　the　computational　burden　and　successfully　estimate　solar　radiation　with　an　accuracy　of　0.91.　Also,　a　quadrant-based　segmentation　method　was　developed　to　classify　rooftop　types　with　an　overall　accuracy　of　83.63%,　which　helps　better　calculate　solar　energy　utilization　cost.　Finally,　we　assumed　PhotoVoltaic　conversion　(PV)　systems　or　Domestic　Solar　Water　Heater　(DSWH)　systems　are　installed　on　all　suitable　rooftops　and　analyzed　their　utilization　cost,　as　well　as　economic　and　ecological　benefits.　The　results　show　that　165,529　and　52,131　buildings　are　suitable　for　PV　and　DSWH　systems,　respectively.　PV　system＇s　average　payback　time　is　about　five　years　with　reducing　standard　coal　consumption　of　1.85　×　1011kg.　The　DSWH　system　requires　three　more　years　to　balance　the　cost　and　benefits.　The　criteria　of　further　solar　energy　planning　are　also　concluded　from　three　perspectives:　building　height　and　density,　and　rooftop　structure.　©　2022　Elsevier　Ltd