Research And Application On Reschedule Technology For Segmented Group Assembly-welding Process In Shipbuilding

Segmented group is the smallest combined unit for ship building,with complex shape and great manufacturing process variation.Its assemblywelding process requires the coordination of task sequences,workers,equipment,parts and working area,and it can be carried out normally only when “Man,Machine,Material,Method,Environment” are all available.During the actual assembly-welding process,many abnormal conditions,such as parts delivery delays and equipment failures occur,resulting in the failure of the initial plan implementation.The current segmented group assemblywelding unit lacks effective monitoring methods and scientific exception handling mechanism,the exception is reported level by level after being discovered,which is slow and inefficient,leading to many bad consequence such as working confusion,low utilization of the resources,delayed progress and extension of the construction period.Therefore,this thesis will carry out research on the re-scheduling technology of segmented group assemblywelding operations,regulate the initial plan affected by abnormal events,reduce the negative impact of abnormal events and improve assembly-welding efficiency.First of all,this thesis constructs a data model for segmented group assembly-welding.According to the segmented group assembly-welding process,the characteristics is extracted and the composition of the assemblywelding data is analyzed.Then combined with the needs of rescheduling research,a segmented group assembly-welding data acquisition system based on the Internet of Things is built.The further analysis for assembly-welding data is taken and the hierarchical relationships are sorted out.With the objectoriented method,the assembly-welding data model is constructed with three aspects of process,planning and execution.The data model describes all process elements with all attributes during the full assembly-welding process,providing data foundation for follow-up research.Then,this thesis studies the segmented group assembly-welding reschedule problem.Based on the analysis of assembly-welding operation process,considering sequence constraints,worker resource constraints and working area constraints,with the goal of minimizing the deviation from the original planned completion time,the mathematical reschedule mode under the constraints of time and space is constructed.The genetic algorithm is used to solve it,with the designed multi-chromosomal encoding and decoding methods that are in line with the characteristics of the assembly-welding operation and the improved area configuration algorithm,the starting time of tasks,the allocation of workers and the configure way of segmented group are adjusted.The algorithm is tested with actual engineering case data,and the results show that the difference in completion time before and after rescheduling is about 2%,which greatly eliminates impact of abnormal events.Finally,this thesis designs and develops a re-scheduling system for the segmented group assembly-welding unit.Based on the assembly-welding data model and the rescheduling algorithm,the rescheduling system is developed.The system can be divided into execution part and control part,contains functional modules of resource management,task monitoring and rescheduling.The execution part is mainly responsible for data collection and task execution,and the control part is for data analysis and decision generation.The combination of the two can realize real-time monitoring of the assembly and welding process,timely identification and response to exception,which can improve the efficiency of the segmented group assembly welding process.