The　nonlinear　relationship　between　the　built　environment　and　active　travel　among　older　adults　within　the　inner-city　context　has　been　widely　researched.　However,　limited　attempts　have　been　made　to　explore　the　demographic　and　spatial　heterogeneity　of　the　built　environment＇s　effects　in　dense　urban　or　peripheral　urban　contexts.　With　the　outward　urban　expansion　brought　about　by　rapid　suburban　growth　in　Chinese　cities,　there　have　been　dynamic　built　environment,　service,　and　socioeconomic　composition　changes　in　the　suburbs.　Using　data　from　Xiamen,　China,　this　study　employs　a　BO-LightGBM　method　to　fill　this　gap　among　different　cohorts　of　the　aging　population　(different　personal　statuses　and　family　relationships)　within　two　contexts,　and　the　SHAP　method　is　adopted　to　interpret　the　models　with　distinctive　spatial　patterns　of　built　environment　differentiation.　Our　results　show　that　differentiation　in　the　threshold　values　and　gradients　exists　between　built　environment　associations　and　the　elderly＇s　active　travel　in　Inner　Xiamen　and　Outer　Xiamen.　The　contribution　of　built　environment　variables　to　impact　older　adult＇s　active　travel　is　related　to　the　regional　context　as　well　as　demographic　characteristics.　These　results　offer　nuanced　and　contextualized　built　environment　interventions　for　building　age-friendly　cities.　©　2023　Elsevier　Ltd