Multiple　tower　crane　layout　planning　(MTCLP)　refers　to　the　selection　of　the　types,　quantities,　and　positions　of　tower　cranes　and　material　storage　positions.　In　a　construction　project,　choosing　tower　cranes　with　a　larger　covering　radius　and　lifting　capacity　can　reduce　the　transportation　time　of　materials.　However,　it　can　also　simultaneously　raise　the　rental　cost　and　the　risk　of　crane　crashes.　Determining　an　optimal　multiple　tower　crane　layout　plan　to　achieve　high　construction　efficiency　with　low　cost　is　still　a　challenge.　Mathematically,　MTCLP　is　a　complex　combinatorial　problem　controlled　by　a　number　of　variables,　such　as　site　configuration,　building　layout,　material　storage　positions,　and　workload.　To　precisely　describe　irregular　shaped　sites　and　buildings,　this　research　partitioned　the　site　into　unit　cells　and　proposed　a　cell-based　optimization　model.　For　the　optimal　crane　layout　plan,　the　problem　was　formulated　as　a　mixed　integer　linear　problem　(MILP).　The　model＇s　mathematical　constraints　comprised　coverage　of　the　given　positions,　the　crane＇s　safety　distance,　sufficient　crane　capacity,　and　the　utilization　ratio.　The　objectives　were　to　increase　the　crane　coverage　ratio　while　reducing　the　cost　and　the　working　areas　overlapped　by　more　than　two　cranes.　Two　case　studies　are　presented　to　demonstrate　the　effectiveness　of　the　model.　The　first　construction　project　consisted　of　seven　buildings　with　irregular　shapes.　By　strictly　forbidding　the　areas　to　be　overlapped　by　more　than　three　tower　cranes,　the　model　successfully　obtained　the　optimal　crane　layout　plan　using　the　Gurobi　Optimizer.　Compared　with　the　contractor＇s　layout　plan,　the　optimal　plan　reduced　the　rental　cost　by　10.7%　and　the　area　overlapped　by　more　than　three　tower　cranes　by　4.56%.　In　the　second　project,　alternative　plans　were　identified　by　the　Gurobi　Optimizer　and　e-constraint　method　to　enhance　confidence　in　selecting　the　ultimate　crane　layout　plan.　The　optimization　procedure　involved　(1)　optimizing　the　layout　plan　with　minimum　crane　cost　and　maximum　coverage　ratio,　(2)　identifying　all　possible　layouts　as　the　Pareto　fronts　when　the　coverage　ratio　could　not　be　raised　without　increasing　the　crane　cost,　and　(3)　minimizing　the　areas　of　crane　overlap　covered　by　more　than　three　tower　cranes.　The　effectiveness　of　the　proposed　model　is　also　proved　by　comparing　the　differences　in　layout　plans　caused　by　＂regular＂　and　＂urgent＂　lifting　demands.