Project 3: Assembly and Joining Sequence Optimization

​Project Leader: Kristina Wärmefjord (Research group: Geometry Assurance & Robust Design)

Academic staff

RG: Geometry and Motion Planning (GMP)
Johan Carlson                         
Robert Bohlin
Evan Shellshear
Domenico Spensieri
Fredrik Ekstedt
Johan Torstensson
Daniel Segerdahl
Staffan Björkenstam

RG: Geometry Assurance and Robust Design (GA)
Rikard Söderberg
Andreas Dagman
Kristina Wärmefjord, PhD Student
Samuel Lorin, PhD Student

RG: Flexible Automation (FA)
Bengt Lennartson
Nina Sundström, PhD Student 
Oskar Wigström, PhD Student
Maziar Mashai, PhD Student

Industrial partners

Volvo Cars AB 
Alf Andersson
Johan Segeborn, Industrial PhD Student
Christer Gullbrandsson

Saab Automobile
Lennart Malmsköld, Industrial PhD Student
Katarina Billett
Roger Gustafsson
Christian Bromander
Roland Roll
Susanne Boderos
Dan Svensson

Volvo Trucks
Henrik Olsson
Per Kylberg
Josefin Hansen
Ingalill Majkvist

Volvo Aero
Torbjörn Norlander
Egon Aronsson

ABB Corporate Research AB   
Daniel Sirkett
Daniel Szugi

RD&T Technology
Lars Lindkvist

FCC
Johan Carlson
Robert Bohlin
Evan Shellshear
Domenico Spensieri
Fredrik Ekstedt
Johan Torstensson
Daniel Segerdahl
Staffan Björkenstam

Background

Design and production involve a number of complex sequences determining the order in which tasks and operations are to be carried out. Optimization of sequences for design activities, assembly, welding, sealing, painting etc. will enable production with highly increased equipment utilization, throughput and quality.
 
An important activity within product and process development is to optimize assembly and joining sequences. The assembly and joining sequence affects geometrical quality, feasibility and cycle time. Geometrical quality since different sequences allow, by design of locating and support schemes, for different levels of control of the variation propagation in the assembly system.
 
Feasibility and cycle time since the sequence influences the dynamic packing/path planning process where there is a need to verify that all parts and tools have a collision-free assembly motion.

Goal

The goal of this project was to develop strategies, methods and tools to support assembly sequence optimization with respect to geometrical quality, feasibility, cycle time, and motion complexity. The project addressed the following research questions (slightly modified from Operational Plan Stage 2, the new RQs 3:11–3:13 replace RQ 3:8–3:9):

  • RQ 3:1
    How can all possible assembly sequences and subassemblies be generated?
  • RQ 3:2
    What processes narrow down the set of possible assembly and disassembly sequences?
  • RQ 3:3
    How can each assembly sequence be optimized with respect to geometrical quality?
  • RQ 3:4
    How can each assembly sequence be verified to be feasible?
  • RQ 3:5
    How can the complexity of an assembly motion be measured?
  • RQ 3:6
    How can the optimal fixturing and clamping strategy be determined? 24 Wingquist Laboratory VINN Excellence Centre Evaluation Report 2012-01-31
  • RQ 3:7
    How can the optimal welding strategy be determined?
  • RQ 3:10
    How can the assembly and joining sequences be optimized with regard to feasibility, cycle time and quality
  • RQ 3:11
    Generic framework for sequence modeling and planning for multi-product assembly station including logic and transportation flows (RG: FA/GMP)
  • RQ 3:12
    How can assembly positions be determined automatically (static packing)? (RG: GMP)
  • RQ 3:13
    How can inspection strategies and sequences be optimized? (RG: GMP/GA)

Project realization

The project results will support the product design and assembly system design in the concept phase of assembled products. The project was carried out in close cooperation with our industrial partners and also involved people from these companies. Case studies were carried out on the basis of problems from the industrial partners.

Summary of results

The project has resulted in a number of methods and algorithms for sequence optimization and simulation. By taking sequences into account, resources can be saved and quality improved. Several of the algorithms are implemented and used in industry.
 
Among the results, a new process for efficient inspection preparation and automatic off-line programming can be mentioned. Methods for solving different variants of the traveling salesman problem (TSP) were combined with path planning in order to find fast and collision-free paths for coordinate-measuring machines. The DMIS code can be automatically generated.
 
Further, the project has resulted in new methods and algorithms for automatic line balancing and path planning of multi-robot stations. The results show 25 percent better equipment utilization and a reduction in engineering time from months to days.
 
Another result from this project is a method for including spot welding sequences in variation simulation. Each executed spot weld leads to forces on both the parts to be joined. By accumulating all forces from fixturing and from applying spot welds, the springback of the final assembly can be calculated.
 
During Stage 2, three new research questions (RQ3:11 -3:13) were formulated. Research questions RQ3:1- 3:4 are mainly addressed in the publications 3:13-3:14. RQ3:1 and RQ3:3 were also addressed during Stage 1 and they were finished during the first year of Stage 2. Research question RQ3:5 are mainly addressed in the publications 3:1-3:4. Research questions RQ3:6-3:7 are mainly addressed in the publications 3:7-3:8, 3:17 and 3:22-3:24. Research question RQ3:10-3:11 are mainly addressed in the publications 3:5-3:6, 3:11, 3:12, 3:16 and 3:18-3:19. Research question RQ3:13 are mainly addressed in the publications 3:20-3:21.

Implementations

 1. Spot-welding sequence simulation


Image: Simulation vs Inspection
Short description/results:
A method for including spot welding sequence in variation simulation was validated on an industrial case study using inspection data. Methods for optimization of spot welding sequence were tested and evaluated on seven industrial case studies [3:23-3:24].
 
Implementation:
A module for spot-welding sequence simulation is included in the commercial software RD&T for variation simulation, used at Volvo Cars.

2. Efficient geometry inspection and off-line programming


Image: Inspection in CMM
Short description/results:
Methods and algorithms allowing easy inspection preparation and automatic generation of collision-free and cycle time-efficient DMIS code [ongoing publication].
 
Implementation:
The method has been implemented in the commercial software RD&T for geometry assurance and IPS for path planning. The method has successfully been put into full production work at Volvo Cars and tested at Saab Automobile.

3. Automatic path planning and line balancing


Image: Load Balancing
Short description/results:
A world-first automatic method for load balancing of welds in robot lines has been developed and tested in joint effort with the FFI project Path Planning and Line Balancing. Application of this method indicates 25 percent better equipment utilization, and 75 percent reduction of off-line programming and commissioning costs [3:12, 3:15].
 
Implementation:
The method has been implemented in the commercial software IPS for path planning and has been awarded the Volvo Cars Technology Award 2011.

4. Disassembly algorithms used in production


Image: Disassembly Planning
Short description/results:
Methods and algorithms supporting vehicle services by calculating collision-free disassembly paths and sequences [3:13].
 
Implementation:
The method has been implemented in the commercial software IPS for path planning. The disassembly method has successfully been put into production work at Volvo Parts and Volvo Cars.

5. Optimization of the geometrical robustness


Image: Robustness Optimization
Short description/results:
Methods for optimization of the geometrical robustness, with respect to both overall robustness and specified critical dimensions, were applied to three components of an aircraft engine. This resulted in suggestions of improved locating schemes, suppressing part and fixture variation [3:17].
 
Implementation:
Methods for optimization of robustness are included in commercial software RD&T for variation simulation, in use at Volvo Aero.

Publication and Presentation Activity

3:1 Bohlin R., Delfs, N., Hanson, L., Högberg, D., Carlson, J.S, 2011,”Unified solution of manikin physics and positioning. Exterior root by introduction of extra parameters”, First International Symposium on Digital Human Modelling, Lyon, France.
 
3:2 Carlson, J. S., Mårdberg, P., Bohlin, R., Delfs, N., Gustafsson, S., 2012, “Using Ergonomic Criterias to Adaptively Define Test Manikins for Design Problems”, 4th International Conference on Applied Human Factors and Ergonomics (AHFE 2012), California, USA.
 
3:3 Delfs, N, Bohlin, R., Mårdberg, P., Gustafsson, S., Carlson, J. S., 2012, “Comparison of Algorithms for Automatic Creation of Virtual Manikin Motions”, 4th International Conference on Applied Human Factors and Ergonomics (AHFE 2012), California, USA.
 
3:4 Gustafsson, S., Tafuri, S., Mårdberg, P., Delfs, N., Carlson, J. S., 2012, “Automatic Fitting of a 3D Character to a Computer Manikin”, 4th International Conference on Applied Human Factors and Ergonomics (AHFE 2012), California, USA.
 
3:5 Magnusson, P., Sundström, N., Bengtsson, K., Lennartson, B., Falkman, P., Fabian, M., 2011, “Planning transport sequences for flexible manufacturing systems”, Proc. 18th IFAC World Congress, Milan, Italy.
 
3:6 Mashaei, M., Lennartson, B. , Sannehed, F. , Abbestam, G., 2010, “Optimal Design of a Decoupled Multiple-Loop Pallet System for Cyclic Flexible Manufacturing Plants”. Proc. 6th IEEE International Conference on Automation Science and Engineering (CASE 2010), Toronto, Canada.
 
3:7 Pahkamaa, A., Wärmefjord, K., Karlsson, L., Söderberg, R., Goldak, J., 2010, “Combining variation simulation with welding simulation for prediction of deformation and residual stress”, Proc. of ASME 2010 International Mechanical Engineering Congress & Exposition, Vancouver, Canada.
 
3:8 Pahkaama, A., Wärmefjord, K., Karlsson, L., Söderberg, R, Goldak, J., 2012, ”Combining Variation Simulation with Welding Simulation for Prediction of Deformation and Geometrical Variation of a Final Assembly”, accepted for publication in Journal of Computing & Information Science in Engineering.
 
3:9 Segeborn, J., Carlsson, A., Carlson, J. S., Söderberg, R., 2009, “A Chronological Framework for Virtual Sheet Metal Assembly Design”, Proc. of the 11th CIRP International Conference on Computer Aided Tolerancing, Annecy, France.
 
3:10 Segeborn, J., Carlsson, A., Carlson, J. S., Söderberg, R., 2010, “A Chronological Framework for Virtual Sheet Metal Assembly Design”, Product Lifecycle Management: Geometric Variations, ISTE Ltd, Wiley, London, pp. 175-190.
 
3.11 Segeborn, J., Carlson, J. S., Wärmefjord, K., Söderberg, R., 2011,”Evaluating Genetic Algorithms on Welding Sequence Optimization with Respect to Dimensional Variation and Cycle Time”, Proc. of ASME IDETC2011, Washington, D. C., USA.
 
3.12 Segeborn, J., Segerdahl, D., Ekstedt, F., Carlson, J. S., Carlsson, A. and Söderberg, R., 2011, “A Generalized Method for Weld Load Balancing in Multi Station Sheet Metal Assembly Lines”, Proc. of the ASME International Mechanical Engineering Congress & Exposition, IMECE2011, Denver, Colorado, USA.
 
3.13 Spensieri, D., Carlson, J. S., Lindkvist, L., Bohlin, R., Söderberg, 2009, “A method to optimize geometrical quality and motion feasibility of assembly sequences”, Proc. of 15th CIRP International Conference on Computer Aided Tolerancing, Annecy, France.
 
3:14 Spensieri, D., Carlson, J. S., Lindkvist, L., Bohlin, R., Söderberg, R., 2010, “A method to optimize geometrical quality and motion feasibility of assembly sequences”, Product Lifecycle Management: Geometric Variations, edited by M. Giordano, L. Mathieu, F. Villeneuve ISTE Ltd and John Wiley&Sons, Inc.
 
3:15 Spensieri, D., Ekstedt, F., Torstensson, J., Bohlin, R., Carlson, J. S., 2010, “Throughput Maximization by Balancing, Sequencing and Coordinating Motions of Operations in Multi-Robot Stations”, Proc. of NordDesign2010 International Conference on Methods and Tools for Product and Production Development, Gothenburg, Sweden.
 
3:16 Thorstensson, C., Kanthabhabhajeya, S., Lennartson, B., Falkman, P. 2011, ”Optimization of Discrete Event Systems using Extended Finite Automata and Mixed-Integer Nonlinear Programming”, Proc. of 18th IFAC World Congress, Milan, Italy.
 
3:17 Vallhagen, J., Lööf, J., Söderberg, R., Wärmefjord, K., 2011, “Robustness in aerospace components manufacturing and fabrication – a case study”. Proc. of the International Association of Airbreathing Engines Conference, Gothenburg, Sweden.
 
3:18 Vergnano, A., Thorstensson, C., Lennartson, B., Falkman, P., Pellicciar, M., Leali, F., Biller, S., 2012, “Modeling and optimization of energy consumption in cooperative multi-robot systems”, Accepted for publication in IEEE Transactions on Automation Science and Engineering.
 
3:19 Vergnano, A., Thorstensson, C., Lennartson, B., Falkman, P., Pellicciar, M., Yuan, C., Biller, S., Leali, F., 2010, “Embedding detailed robot energy optimization into high-level scheduling”, Proc. 6th IEEE International Conference on Automation Science and Engineering (CASE 2010), Toronto, Canada.
 
3:20 Wärmefjord, K., Carlson, J. S., Söderberg, R., 2010, “A Measure of the Information Loss for Inspection Point Reduction”, Journal of Manufacturing Science and Engineering, 131 (5), pp 051017-1- 051017-6.
 
3:21 Wärmefjord, K., Carlson, J. S., Söderberg, R., 2010, “An investigation of the effect of sample size on geometrical inspection point reduction using cluster analysis”, CIRP Journal of Manufacturing Science and Technology, 3 (3) pp. 227-235.
 
3:22 Wärmefjord, K., Söderberg, R., Carlson, J., 2010, “Including fixture repeatability in variation simulation”, Proc. of ASME 2010 International Mechanical Engineering Congress & Exposition, Vancouver, Canada.
 
3:23 Wärmefjord, K., Söderberg, R., Lindkvist, L., 2010, “Strategies for Optimization of Spot Welding Sequence With Respect to Geometrical Variation in Sheet Metal Assemblies”, Proc. of ASME 2010 International Mechanical Engineering Congress & Exposition, Vancouver, Canada.
 
3:24 Wärmefjord, K., Söderberg, R., Lindkvist, L., 2010, “Variation simulation of spot welding sequence for sheet metal assemblies”, Proc. of Nord- Design2010 International Conference on Methods and Tools for Product and Production Development, Gothenburg, Sweden.

Theses

 
3:25 Bengtsson, C., 2011, “Efficient and Accurate Computation of Rectangular Swept Spheres Using Minimum Bounding Ellipse Algorithms”, Master’s thesis, University of Gothenburg, Gothenburg, Sweden.
 
3:26 Delfs, N.,2011, “Manikin in Time; Development of a Virtual Manikin with Inverse Kinematics and Comfort”, Master’s thesis, University of Gothenburg, Gothenburg, Sweden.
 
3:27 Eek, G., Eriksson C., 2009, “Effective methods for solving the balanced and synchronised multiple TSP using genetic algorithms”, Master thesis, University of Gothenburg, Gothenburg, Sweden.
 
3:28 Jönsson, V., Ustyan, T., 2011, “Implementation of a Generic Virtual Robot Controller”, Master thesis, Chalmers University of Technology, Gothenburg, Sweden.
 
3:29 Magnusson, P., 2009, “Implementation of an interconnection between soft PLC and simulation software”, Master thesis, Chalmers University of Technology, Gothenburg, Sweden.
 
3:30 Segeborn, J., 2011, “Cost-effective Sheet Metal Assembly by Automatic Path Planning and Line Balancing, Integrated with Dimensional Variation Analysis”, Doctoral Thesis, Department of Product and Production Development, Chalmers University of Technology, Gothenburg, Sweden.
 
3:31 Wärmefjord, K., 2011, “Variation Control in Virtual Product Realization - A Statistical Approach”, Doctoral Thesis, Department of Product and Production Development, Chalmers University of Technology, Gothenburg, Sweden.

Published: Wed 28 May 2014.