The　residual　strain　in　asphalt　pavement　changes　greatly　from　year　to　year,　in　different　seasons　and　through　different　periods　of　the　same　day.　However,　the　residual　strain　cannot　be　measured　directly　and　is　difficult　to　predict　effectively.　Therefore,　in　this　article,　we　propose　a　method　of　monitoring　residual　strain　data　by　continuously　recording　measurements　with　sensors　embedded　in　the　pavement　and　calculating　the　residual　strain.　Considering　the　various　trends　and　dynamic　characteristics　of　the　data,　a　dynamic　prediction　model　was　established　to　precisely　report　the　residual　strain　in　asphalt　pavement　based　on　times-series　approaches.　The　model　is　verified　by　analyzing　predicted　data　recorded　from　Oct.　2015　to　Mar.　2016　on　a　test　road.　The　results　indicate　that　the　dynamic　predictions　of　residual　strain　obtained　using　the　proposed　model　are　accurate,　stable　and　reliable.　©　2017,　The　Editorial　Board　of　Journal　of　Basic　Science　and　Engineering.　All　right　reserved.