PT Dok dan Perkapalan Surabaya (DPS) is one of several strategic BUMN in marine field. With BUMN restructurisation and efficiency program from Government, PT DPS is liable to improve its production system efficiency. Many problems that faced by PT DPS to improve its production system efficiency are : (1)There is waste work shop such as work in process, waiting time, and flow time at fabrication process in PT DPS, (2) Because waste work shop is still big, shop production process efficiency has not reached maximally yet, (3)Layout and facilities of PT DPS today is already suitable with concept of Product Oriented Work Breakdown Structure (PBWS), but it is limited at simple construction form. At this research, there is ship production process design with Process Lane Construction and Zone Outfitting method or it is known with Group Technology Layout (Manufacturing Cell System) term in order to solve inefficiency problem at ship production process in PT DPS. Design process is performed by integrating Simulation method, Taguchi, Response Surface Methodology, and Analytical Hierarchy Process (AHP). Simulation method is used to model designed manufacturing cell system, meanwhile Taguchi method and Response Surface Methodology are used for experiment and optimization at manufacturing system simulation model. In order to measure the performanceof designed manufacturing cell system, multivariate measurement method is used, that is by measuring work in process, waiting time, and flow time with smaller the better criteria. Multi response becomes single response by using Analytical Hierarchy Process (AHP) method. From research result with Taguchi Method and Response Surface Methodology know that in fabrication stage optimal layout is second tipe, optimal set up time is 40 minutes, optimal lot size is 16 tons, optimal loading interval is 400 minutes, optimal demand stability is 92.5 %, in subassembly stage optimal layout is first tipe, optimal set up time is 40 minutes, optimal lotsize is 10 tons, optimal loading interval is 400 minutes, optimal demand stability is 92.5 %, in assembly stage optimal layout is first tipe, optimal set up time is 40 minutes, optimal lotsize is 6tons, optimal loading interval is 400 minutes, optimal demand stability is 92.5 %, in erection stage optimal layout is second tipe, optimal set up time is 40 minutes, optimal lot size is 6 tons, optimal loading interval is 480 minutes, optimal demand stability is 92.5 %.