The　primary　feature　of　Industry　4.0　is　MPP　(mass　personalization　production),　which　requires　that　consumers＇　individual　requests　are　met　in　large-scale　production.　Under　MPP,　there　is　a　multitude　of　subtasks　decomposed　from　production　tasks　that　are　derived　from　individualized　requests,　and　allocating　these　subtasks　properly　brings　high　economic　benefits.　However,　existing　approaches　to　achieve　MPP,　such　as　cloud　manufacturing　and　social　manufacturing,　generally　can　not　provide　customers　with　deep　participation　in　the　entire　production　cycle,　or　respond　to　consumers＇　modification　needs　by　a　triggered　mechanism.　Besides,　some　methods　are　of　centralized　architecture,　which　is　vulnerable　to　single　point　error　and　with　large　cloud　load　that　is　not　conducive　to　quickly　responding　to　consumers＇　dynamic　demand　changes.　Therefore,　this　paper　proposes　a　dynamic　edge-cloud　manufacturing　mode　for　MPP,　which　can　make　subtask　allocation　with　high　economic　benefit　through　distributed　computing　and　implementing　modifications　of　alternating　direction　method　of　multiplier　(ADMM)　algorithm.　Also,　it　proposes　an　original　improved　ADMM　algorithm,　named　Relaxation-Based　ADMM　algorithm,　to　increase　the　optimization　speed　in　large-scale　cases.　The　experimental　results　show　that　the　proposed　method　generally　obtains　a　superior　solution　under　a　certain　iteration　count.(c)　2023　The　Authors.　Published　by　Elsevier　B.V.　on　behalf　of　King　Saud　University.　This　is　an　open　access　article　under　the　CC　BY-NC-ND　license　(http://creativecommons.org/licenses/by-nc-nd/4.0/).