In　this　paper,　we　present　a　lightweight,　integrated　LiDAR　SLAM　system　designed　for　high　efficiency　and　robust　loop　closure　detection.　Our　evaluation　focuses　on　LiDAR-based　platforms　with　limited　on-board　computation　capabilities　in　long-distance　and　large-scale　scenarios.　We　identify　that　front-end　only　SLAM　systems　are　prone　to　drift　due　to　accumulated　errors　in　pose　estimation.　To　mitigate　this　issue,　we　propose　two　enhancements:　1)　the　implementation　of　an　incremental　kd　tree　(ikd-tree)　data　structure　for　efficient　map　management,　and　2)　the　utilization　of　Normal　Distribution　Descriptor　(NDD)　for　loop　closure　detection,　combined　with　GTSAM　for　global　optimization,　which　effectively　corrects　the　final　trajectory　of　the　algorithm.　Finally,　we　validate　our　proposed　method　using　the　KITTI　and　MulRan　datasets　and　benchmark　it　against　the　FLOAM　system.　The　experimental　results　reveal　that　our　method　surpasses　FLOAM　in　terms　of　computational　efficiency　by　26.21%　and　53.53%,　and　in　terms　of　accuracy　by　36.88%　and　84.76%　on　the　KITTI　and　MulRan　datasets,　respectively.　These　findings　demonstrate　the　potential　of　our　lightweight　integrated　LiDAR　SLAM　system　to　significantly　improve　both　efficiency　and　accuracy　in　challenging　environments,　making　it　a　valuable　contribution　to　the　field　of　SLAM　research.　©　2024,　The　Author(s),　under　exclusive　license　to　Springer　Nature　Singapore　Pte　Ltd.