1. 1 Helsinki University of Technology Systems Analysis Laboratory Robust Portfolio Modeling in the Development of National Research Priorities Ville Brummer and Ahti Salo Systems Analysis Laboratory Helsinki University of Technology P.O. Box 1100, 02015 TKK, Finland http://www.sal.tkk.fi firstname.lastname@tkk.fi

2. Helsinki University of Technology Systems Analysis Laboratory 2EURO 2006 XXI - Iceland RPM – Robust Portfolio Modelling

3. Helsinki University of Technology Systems Analysis Laboratory 3EURO 2006 XXI - Iceland Multi-Criteria Project Portfolio Selection n Choose a portfolio of projects from a large set of proposals –Projects evaluated with regard to multiple criteria –Resource and other portfolio constraints n One of the most important areas of MCDM applications –R&D portfolio selection (Golabi, Kirkwood and Sicherman, 1981; Stummer and Heidenberger 2003) –Capital budgeting (Kleinmuntz and Kleinmuntz, 1999) –Strategic product portfolio selection (Lindstedt, Liesiö and Salo, 2006) –Evaluation of innovation programss (Salo, Mild, Pentikäinen, 2006) –Selection of forest sites for conservation (Punkka, 2006) –Road asset management (Mild, 2006)

4. Helsinki University of Technology Systems Analysis Laboratory 4EURO 2006 XXI - Iceland Robust Portfolio Modelling (RPM) n Liesiö, Mild, Salo (2006). Preference Programming for Robust Portfolio Modelling and Project Selection, forthcoming in EJOR n Projects n Evaluation with regard to multiple criteria –Score of project with regard to the i-th criterion –Criterion weights n Additive representation of project value

5. Helsinki University of Technology Systems Analysis Laboratory 5EURO 2006 XXI - Iceland n Portfolio p = a subset of projects n Portfolio value = sum of its projects’ values (Golabi et al. 1981) n Feasible portfolios satisfy linear constraints –E.g., budget constrains n Maximize portfolio value –Zero-one linear programming problem (ZOLP) Project Portfolios

6. Helsinki University of Technology Systems Analysis Laboratory 6EURO 2006 XXI - Iceland n Elicitation of complete information (point estimates) on weights and scores may be costly or even impossible n Weights constrained by the DMs preference statements –Several kinds of preference statements impose linear constraints on weights –(Incomplete) rank-orderings on criteria (cf., Salo and Punkka, 2005) –Interval SMART/SWING (Mustajoki et al., 2005) n Intervals of project-specific scores –Lower and upper bounds on criterion-specific scores of each project n Information set Incomplete Information in Portfolio Selection

7. Helsinki University of Technology Systems Analysis Laboratory 7EURO 2006 XXI - Iceland n Portfolio p dominates p’ on S, denoted by, if n Portfolio p’ can be discarded because p yields higher value! n Non-dominated portfolios (NDP) –Restrict attention to NDPs only –All NDPs computed by a dedicated dynamic programming algorithm –Multi-Objective Zero-One LP (MOZOLP) problem with interval-valued objective function coefficients Dominance Concept for Portfolios

8. Helsinki University of Technology Systems Analysis Laboratory 8EURO 2006 XXI - Iceland Recommendations at the Portfolio Level n Core Index of a project, –Share of non-dominated portfolios (NDP) that contains n Core projects, i.e., can be surely recommended –Would belong to all NDP even if additional information is acquired n Exterior projects, i.e., can be safely rejected –Cannot enter any NDP even with additional information n Borderline projects, i.e., need further analysis –Negotiation / iteration zone for augmenting the set of core projects –Narrower score intervals help reduce the set of borderline projects

9. Helsinki University of Technology Systems Analysis Laboratory 9EURO 2006 XXI - Iceland Non-dominated Portfolios and Core Index n Non-dominated portfolios (NDP) –No other feasible portfolio gives higher overall value with all feasible weights and scores n Project’s Core Index (CI) –Core proj. are included in all NDP (CI=1) –Exterior proj. not included in any NDP (CI=0) –Borderline proj. included in some NDP (0<CI<1) 6 3 10 2 1 7 4 5 8 9 A B C A Overall value at extreme point 1 B C E D B Overall value at extreme point 2 C A D E D Overall value at extreme point 3 A E C B * ** * * *

10. Helsinki University of Technology Systems Analysis Laboratory 10EURO 2006 XXI - Iceland Core Index Analysis n Core Index is a performance of a measure which accounts for –Incomplete information on weights and scores –Project cost and competing proposals –Budget and other feasibility constraints n Helps classify projects Hundreds of projects Multiple criteria Portfolio-level constraints Borderline proj → focus Exterior proj → reject Core projects → accept Compute non-dominated portfolios Incomplete information

11. Helsinki University of Technology Systems Analysis Laboratory 11EURO 2006 XXI - Iceland Case: Development of National Research Priorities

12. Helsinki University of Technology Systems Analysis Laboratory 12EURO 2006 XXI - Iceland Forest-Based Sector Technology Platform (FTP) n One of the over 30 European Technology Platforms –Coordination of industry-lead European R&D activities –Establishment of the European Research Area (ERA) n This particular Technology Platform initiated by –European Confederation of Woodworking Industries –Confederation of European Forest Owners –Confederation of European Paper Industries n Over 30 countries involved –Launched in 2003 –Long-term perspective (2030) –Development of the Strategic Research Agenda (SRA) in 2005 in member states and at the European level

13. Helsinki University of Technology Systems Analysis Laboratory 13EURO 2006 XXI - Iceland n Step 1: Each country was requested to identify 10 -15 most relevant research themes in view of national priorities Strategic Research Agenda (SRA) for the FTP

14. Helsinki University of Technology Systems Analysis Laboratory 14EURO 2006 XXI - Iceland Challenges ŒNew policy instrument  No established approaches A very wide range of issues to be covered –Many stakeholder groups (e.g., pulp and paper industry, bioenergy, forestry) –Long time scale  considerable uncertainties ŽTight timetable –Only 7 weeks  Need for a structured decision support process Multiple interfaces to other policy processes –E.g., preparation of Framework Program (FP7) in Europe n Forest sector is a key industry in Finland –24 % of exports

15. Helsinki University of Technology Systems Analysis Laboratory 15EURO 2006 XXI - Iceland Finnish Case: National SRA Process for the FTP n Systematic process to engage Finnish key stakeholders –Development of the national SRA –Linked explicitly to the Vision 2030 document at the European level n Five value chains –Forestry, Wood Products, Pulp and Paper, Bio Energy, Specialties/ New Businesses –Independent but interrelated process for each value chain n Identification and assessment of research themes –Internet questionnaires – MCDM analysis - interactive decision workshops n Synthetisation of national results at the end of the process

16. Helsinki University of Technology Systems Analysis Laboratory 16EURO 2006 XXI - Iceland Participants and Roles n Steering Group –Coordinators and selected key persons (~ 10 people) n Coordinators –Chairs of national value chain Working Groups (5 people) n TKK Group –Research team of Prof. Ahti Salo at the Systems Analysis Laboratory / TKK n Respondents –20-30 participants within each value chain n Referees –6-10 participants within each value chain

17. Helsinki University of Technology Systems Analysis Laboratory 17EURO 2006 XXI - Iceland Process Design Process stepsWeeksKey participants I Step: Internet-based solicitation of research themes 1-2Respondents II Step: Internet-based assessment of research themes 3-4Referees III Step: Multi-criteria analysis of research themes 4-5TKK group IV Step: Value chain workshops for the formulation of relevant research areas 5-6Value Chain Coordinators and invited Respondents, Referees and other experts VI Step: Steering Group workshop for the formulation of Finnish SRA priorities 7Steering Group

18. Helsinki University of Technology Systems Analysis Laboratory 18EURO 2006 XXI - Iceland Task 1: Solicitation of Research Themes n Timetable: April 27 – May 8 n Participants: 20 -30 Respondents / Value chain n Task: In each value chain, respondents proposed research themes with the Opinions-Online decision support tool n Result: Total 146 research themes n Task 1 Task 1 n Example Example

19. Helsinki University of Technology Systems Analysis Laboratory 19EURO 2006 XXI - Iceland Task 2: Assessment of Research Themes n Timetable: Mid-may n Participants: 6 -10 Referees / Value chain n Task: In each value chain, referees assessed research themes with Opinions-Online© n Result: Numerical assessment of research themes with regard to different criteria n Task 2 Task 2

20. Helsinki University of Technology Systems Analysis Laboratory 20EURO 2006 XXI - Iceland Task 3: Analysis on the Results n Timetable: Mid-may n Participants: Research group at TKK n Task: TKK group analysed the results using RPM- methodology –”Research theme” as ”project” treated as equal unit –Scores defined as average of criterion specific evaluations –Highlight 7 the most interesting themes from the whole set (Budget: ) n Result: Shortlist of ’the most interesting’ themes on each value chain n Example Example

21. Helsinki University of Technology Systems Analysis Laboratory 21EURO 2006 XXI - Iceland Task 4: Value Chain Workshops n Timetable: May 23 -31 n Participants: Value chain working groups –Selected respondents, referees and other experts n Task: Value chain Working Groups discussed on the results and identified most relevant research themes. n Result: 3-7 the most relevant themes from each value chain

22. Helsinki University of Technology Systems Analysis Laboratory 22EURO 2006 XXI - Iceland Task 5: SRA Workshops n Timetable: June 8 n Participants: SRA steering group (includes value chain coordinators) n Task: Based on the results from previous tasks and especially from value chain workshops, SRA steering group identified the most relevant 15 research themes n Result: the most relevant research themes –Taken forward to European level

23. Helsinki University of Technology Systems Analysis Laboratory 23EURO 2006 XXI - Iceland Conclusions n Systematic way of organising foresight processes –Permits extensive stakeholder participation even with tight schedules –Is transparent in terms of methodology –Supports workshop discussions through MCDA analyses n Considerations –Formal MCDA inputs are helpful but need to be complemented –Supports discussions by synthesizing results based on Core Index values –Makes it possible to consider multiple perspectives (criteria & their weights) n Applicable in several other contexts, too