The　surface　rupture　triggered　by　strong　earthquakes　will　extensively　damage　buildings　on　the　ground.　It　is　significant　for　choosing　construction　sites　to　forecast　accurately　and　calculate　reasonable　building　lien　distance.　The　paper　proposed　a　method　which　aimed　to　forecast　the　surface　rupture　width　under　strong　earthquake　based　on　information　diffusion　methodology.　Firstly,　magnitude　and　the　thickness　of　overburden　soil　were　taken　as　main　evaluation　indexes,　and　then　fuzzy　relation　between　two　evaluation　indexes　and　surface　rupture　zone　width　was　established　based　on　the　information　diffusion　principle　and　fuzzy　approximate　deduction.　Secondly,　cross　validation　was　used　to　evaluate　the　model＇s　stability　and　accuracy,　then　followed　by　a　comparation　with　the　multiple　linear　regression　model,　quadratic　regression　model　and　BP　neural　network　model.　The　results　show　that　information　diffusion　methodology　can　deal　with　nonlinear　relationship　of　every　index.　Finally,　the　validity　and　feasibility　were　proved　by　instance　analysis.