As　an　important　approach　to　avoiding　the　safety　accidents　of　bridges　and　ensuring　the　safety　of　the　public,　bridge　risk　assessments　have　been　widely　concerned　by　lots　of　scholars.　However,　the　existing　models　of　bridge　risk　assessment　mostly　ignored　the　optimization　of　the　value　of　parameters　and　the　number　of　parameters　in　modeling　process,　leading　to　the　failure　of　improving　the　accuracy　of　bridge　risk　assessment.　Therefore,　based　on　the　existing　bridge　risk　assessment　model　using　extended　belief　rule　base　(EBRB),　this　paper　proposes　a　rule　generation　method　and　a　rule　reduction　method　based　on　parameter　optimization　and　data　envelopment　analysis,　then　a　joint　optimization　methods　is　further　proposed　for　optimizing　EBRB.　Finally,　a　commonly　used　and　well-known　dataset　about　bridge　risk　assessment　is　applied　to　validate　the　efficiency　of　the　proposed　model.　In　comparison　with　the　results　of　previous　studies,　the　bridge　risk　assessment　model　based　on　EBRB　with　joint　optimization　has　shown　superior　performance　in　both　improving　the　evaluation　accuracy　and　reducing　the　modeling　complexity.　©　2020,　Editorial　Board　of　Journal　of　Systems　Engineering　Society　of　China.　All　right　reserved.