With　the　ongoing　development　of　China’s　road　infrastructure　and　the　increasing　levels　of　motorization,　the　riding　comfort　of　road　facilities　has　become　a　critical　concern　for　traffic　participants.　The　scientific　and　intelligent　evaluation　of　road　facility　riding　comfort　is　of　great　significance　for　ensuring　their　effective　construction　and　maintenance.　Conventional　riding　comfort　evaluation　has　primarily　relied　on　the　International　Roughness　Index　(IRI),　wherein　higher　IRI　values　correlate　with　poorer　surface　smoothness　and　reduced　driving　comfort.　However,　current　IRI　methods　lack　sophisticated　intelligence　capabilities　and　cannot　achieve　real-time　monitoring　of　either　IRI　or　driving　comfort.　In　response　to　these　challenges,　this　study　introduces　a　self-developed　intelligent　data　collection　device　for　measuring　vibration　acceleration　during　vehicle　operation,　and　establishes　an　innovative　evaluation　method　that　incorporates　vehicle　vibration　acceleration　distribution　characteristics.　Initially,　we　developed　a　preprocessing　protocol　for　vibration　acceleration　data　and　identified　relevant　characteristic　indicators.　This　foundation　enabled　the　creation　of　three　integrated　models:　a　probability　density　model　for　vibration　acceleration　considering　passenger　variability,　a　membership　degree　model　for　vibration　acceleration　comfort,　and　a　comfort　rate　model　incorporating　individual　passenger　differences.　Subsequently,　a　comprehensive　comfort　rate　model　incorporating　vibration　acceleration　distribution　characteristics　was　constructed.　The　study　conducted　driving　tests　were　across　7　road　sections　with　varying　surface　technical　performance,　utilizing　11　vehicle　types　at　different　driving　speeds.　Through　mathematical　statistics　and　stochastic　theory　analysis,　a　Lognormal　distribution　model　of　vehicle　vibration　acceleration　under　mixed　traffic　conditions　was　established.　The　research　results　indicate　that　the　intelligent　data　collection　terminal　exhibits　excellent　working　stability,　as　well　as　high　repeatability　and　reliability　in　data　collection.　The　proposed　evaluation　model　yields　road　surface　comfort　assessments　that　strongly　correlate　with　established　ISO　evaluation　methods　while　offering　superior　discrimination　capabilities　across　various　road　conditions,　thereby　facilitating　optimized　road　maintenance　decision-making　by　relevant　agencies.　Notably,　the　developed　driving　data　acquisition　device　offers　significant　cost　advantages,　with　implementation　expenses　approximately　75%　lower　than　traditional　IRI　detection　methods.　This　research　successfully　achieves　preliminary　intelligent　evaluation　of　road　riding　comfort　through　the　integration　of　self-developed　data　collection　terminal　and　innovative　evaluation　model,　enabling　indirect　intelligent　assessment　of　IRI　values.　The　method　presented　herein　represents　a　significant　advancement　in　road　quality　evaluation　and　maintenance　management,　offering　significant　theoretical　and　practical　contributions　to　the　field.　©　The　Author(s),　under　exclusive　licence　to　Chinese　Society　of　Pavement　Engineering　2025.