NSF Workshop on Logic Control for Manufacturing Systems

NSF Workshop on Logic Control for Manufacturing Systems
University of Michigan, Ann Arbor
June 26-27, 2000
Sponsored by NSF and ERC/RMS
Breakout Session Assignments

The breakout sessions will serve as a forum to discuss relevant issues in logic control for manufacturing systems. The breakout sessions are vitally important as they will be the source of most of the conclusions and recommendations from the workshop.

Each breakout session will consist of a group of Industry members, Academics, Researchers, and Students. At the discretion of the session leader, the group may decide to break into smaller subgroups or may remain whole. The group may address some or all of the listed questions, or may formulate their own list of topics to discuss. The questions are only provided to help get the discussion started.
Each group should begin by appointing a scribe (or scribes) to record the comments. You are strongly encouraged to follow the practice of writing EVERYTHING on the large pads of paper (or overhead transparencies) then taping the pages to the walls around the room as they are filled. This is a time for free flow of ideas rather than debate.
The breakout session leader is asked to present the findings of the group in the reporting sessions following the breakout sessions. Each leader will have 15 minutes to summarize the group discussions. Overhead transparencies and markers will be provided in each breakout session room for use in the reporting session. The LCD projector may also be used.
The breakout session leader, together with the scribe, may also be asked to prepare a written summary of the discussions, which will be used in the final report to NSF.

Breakout session 1: Industry Needs and State-of-the-Art

(Underlined academic is tentative session leader, underlined student is tentative scribe)

A. End-User Needs

I: John Hill (PCC Olofsson), Larry Peterson (Ford), Dave Richart (Cummins), Jerry Yen (GM),

A: Alan Desrochers (RPI), Jin Jiang (Western Ontario), Ratnesh Kumar (Kentucky)

R: Zbigniew Pasek (UM)

S: Cari Bryant (UM), Maurizio Ficocelli (Toronto), Morrison Lucas (UM)

Questions/Topics to start the discussion:

1. Is current functionality of logic control sufficient or does it need to be improved? What new functionality is needed (integration with enterprise management, improved fault handling/diagnostics, ...)

2. Is there support from the end users to move towards different programming languages for logic control?

3. How much access to the control code do end users need?

4. How important is standardization of operator interfaces? how does this affect the choice of logic controller?

5. How important is "Open Architecture" (and how should it be defined)?

6. What is your "wish list" for future logic controllers?

B. Machine Builder Needs

I: Bryan Graham (Lamb), Yonghua Li (Orion), James Mooney (Comau), Bob Stevens (PCC Olofsson)

A: Placid Ferreira (Illinois), Junghsen Lieh (Wright State), Radha Sarma (Toledo)

R: Robert Landers (UM)

S: Dhruba Kalita (UM), Christos Panayiotou (Boston U), Robert Peterson (UM), Alejandro Ramirez-Serrano (Toronto)

Questions/Topics to start the discussion:

1. What are the current obstacles in writing the logic control for manufacturing systems? What could improve the process?

2. Would using different programming languages make the logic control programmers job easier or harder?

3. Are the specifications currently received from end users sufficient? If not, how can they be improved?

4. How important is "Open Architecture" (and how should it be defined)?

5. What is your "wish list" for future logic controllers?

C. State of the Art in Industrial Control

I: Bertil Brandin (Siemens), James Christensen (Rockwell), Aldo Marcuzzi (Nematron), Jon Robinson (PCC Olofsson)

A: Beno Benhabib (Toronto), Christos Cassandras (Boston), Georg Frey (Kaiserslauten)

R: Byung-Kwon Min (UM)

S: Eric Klavins (UM), Feng-Li Lian (UM), Hong Zhang (Wright State)

Questions/Topics to start the discussion:

1. What are the most exciting recent advances in the logic control industry (networking, open architecture, verification tools,...)? How widespread do you believe that they will become? How long will they take to become prevalent?

2. What are the current challenges facing the logic control industry?

3. Do you believe that an open standard will emerge? If so, how could it change the playing field?

4. What role should PC/Windows based controllers play in manufacturing control systems?

5. What is your "wish list" for future logic controllers?

D. Collaboration Avenues

I: Sushil Birla (GM), K. C. Chuang (SupreTech), Andrei Moldovenau (Nematron)

A: Lawrence Holloway (Kentucky), Feng Lin (Wayne State), Mengchu Zhou (NJIT),

R: Euisu Park (UM)

S: Samuel Chan (Toronto), Rami Debouk (UM), Eric Endsley (UM), Iakov Romanovski (McGill)

Questions/Topics to start the discussion:

1. How can industry and universities work together to address problems in logic control?

2. How can industry consortia advance the state of the art in logic control? What participation/support is needed?

3. Are there topics in logic control that could/should be taught in an undergraduate curriculum? In EE, ME, Manufacturing Engineering?

4. What is the correct role for ASME/IEEE/SME committees in this arena? What about NSF?

5. What advances could be made by enhancing industry/university collaboration?

Breakout session 2: Academic State-of-the-Art and Future Research Directions

(Underlined industry is tentative session leader, underlined student is tentative scribe)

A. Research Issues in Logic Control: FSM Framework

I: Sushil Birla (GM), Bertil Brandin (Siemens), Yonghua Li (Orion), Bob Stevens (PCC Olofsson)

A: Beno Benhabib (Toronto), Stephane Lafortune (UM), Feng Lin (Wayne State)

R: Zbigniew Pasek (UM)

S: Rami Debouk (UM), Eric Endsley (UM), Christos Panayiotou (Boston U), Alejandro Ramierz-Serrano (Toronto)

Questions/Topics to start the discussion:

1. What are the most exciting recent developments in logic control using an FSM framework?

2. What are the barriers to implementing research results into industry (lack of communication, academics solving the wrong problems, industrial problems are too complex, speed of computers for solving complex problems, lack of specification for industrial problems, ...)

3. How can these barriers be overcome?

4. What are the best examples of logic control research being implemented in industrial systems? What can we learn from these examples?

5. What are the most important problems that need to be solved in logic control using an FSM framework?

B. Research Issues in Logic Control: Petri Net Framework

I: Bryan Graham (Lamb), John Hill (PCC Olofsson), Andrei Moldovenau (Nematron), Dave Richart (Cummins)

A: Alan Desrochers (RPI), Georg Frey (Kaiserslauten), Mengchu Zhou (NJIT)

R: Euisu Park (UM)

S: Cari Bryant (UM), Maurizio Ficocelli (Toronto), Robert Peterson (UM), Hong Zhang (Wright State)

Questions/Topics to start the discussion:

1. What are the most exciting recent developments in logic control using a Petri net framework?

3. How can these barriers be overcome?

4. What are the best examples of logic control research being implemented in industrial systems? What can we learn from these examples?

5. What are the most important problems that need to be solved in logic control using a Petri net framework?

C. Research Issues in Logic Control: Emerging Frameworks

I: James Christensen (Rockwell), Aldo Marcuzzi (Nematron), Larry Peterson (Ford)

A: Placid Ferreira (Illinois), Jin Jiang (Western Ontario), Lawrence Holloway (Kentucky), Junghsen Lieh (Wright State)

R: Byung-Kwon Min (UM)

S: Dhruba Kalita (UM), Eric Klavins (UM), Morrison Lucas (UM), Iakov Romanovski (McGill)

Questions/Topics to start the discussion:

1. What are the most exciting and promising new frameworks for logic control? (Flow charts, temporal logic, new standards, hybrid (continuous/discrete) control, ....)

3. How can these barriers be overcome?

4. What are the best examples of logic control research being implemented in industrial systems? What can we learn from these examples?

5. What are the most important problems that need to be solved in logic control? How can new or emerging frameworks address these problems better than existing frameworks could?

D. Technology Transfer

I: K. C. Chuang (SupreTech), James Mooney (Comau), Jon Robinson (PCC Olofsson)

A: Christos Cassandras (Boston), Ratnesh Kumar (Kentucky), Radha Sarma (Toledo), Ken Saunders (Penn State)

R: Robert Landers (UM)

S: Samuel Chan (Toronto),Feng-Li Lian (UM), Tae-Sic Yoo (UM)

Questions/Topics to start the discussion:

1. What are the most important barriers to implementing research results into industry (lack of communication, academics solving the wrong problems, industrial problems are too complex, speed of computers for solving complex problems, lack of specification for industrial problems, ...)

2. What can industry do to help overcome these barriers?

3. What can academics do to help overcome these barriers?

4. What are the issues with intellectual property and how can they be addressed?

5. Should academics publish their results freely or patent then? Which makes it more likely for industry to adopt the work?