The　analytic　hierarchy　process　(AHP)　approach　is　applied　to　solving　multi-criteria　decision-making　(MCDM)　problems　in　many　areas.　However,　the　comparison　matrix　(CM)　is　not　＇consistent＇　within　a　limit　of　a　consistency　measure.　An　improved　AHP　(IAHP)　is　proposed　to　improve　CM　consistency　by　using　a　sorting　methodology　to　determine　CM.　A　case　study　is　presented　to　illustrate　the　advantages,　in　terms　of　CMs　consistency　and　information　extraction　effectiveness,　when　IAHP　was　applied　to　risk　identification　during　an　open-cut　subway　construction.　The　distribution　of　the　risks　calculated　by　IAHP　is　in　accordance　with　the　available　statistical　findings.