Transcript The nature of EU funded R&D collaboration networks in the

The nature of EU funded R&D
collaboration networks in the area of
Information Society Technologies
Aimilia Protogerou, Yannis Caloghirou and Evangelos Siokas
Laboratory of Industrial and Energy Economics
National Technical University of Athens
DIME Workshop “Distributed Networks and the Knowledge-based Economy”
GREDEG-DEMOS
Juan les Pins, 10-11 May 2007
1
Aim of the paper

The exploratory study of the structure
and evolution of the research
collaboration networks formed under
the 4th, 5th and 6th EU Framework
Programmes during the period 19942006 in the area of Information Society
Technologies (IST).
2
Exploring EU-funded Research
Joint Ventures


The STEP to RJVs (TSER Project) was the first
attempt to study EU funded research
partnerships.
This study is part of a number of studies that
employ social network analysis to investigate
the formation, structure, and evolution of
networks emerging as a response to EU
policy initiatives.
3
The RJV-IST type of networks



These networks are comprised of firms,
universities, research institutes and
other organizations that get connected
by cooperative contractual agreements.
Exploration networks
Policy-driven networks vs. other
network types studied through patent
citations or technology alliances.
4
Descriptive statistics of the RJVIST database
FP4
FP5
FP6
No of organizations
4,879
7,406
3,879
No of participations
9,302
16,381
10,582
No of projects
1,416
2,206
799
Average duration (months)*
25.65 (8.5)
26.3 (8.82)
32.85 (8.07)
Average budget per project (million €)
2.25 (3.72)
2.68 (2.05)
3.6 (6.45)
Average funding per project (million €)
1.08 (2.17)
1.5 (1.04)
2.2 (3.57)
Average no of participants per project
6.57 (4.26)
7.43 (5.05)
13.24 (9.11)
Average project number per participant
1.9 (3.15)
2.2 (5.09)
2.7 (5.6)
5
Distribution of entity types in the
RJV-IST networks
Entities
Participations
100%
100%
80%
80%
60%
60%
40%
40%
20%
20%
0%
0%
FP4
Education
FP5
Research
Industry
FP6
Other
FP4
Education
FP5
Research
Industry
FP6
Other
6
Returning and new entities by
organization type
(FP5 vs. FP4 and FP6 vs. FP4 +FP5)
100%
90%
FP5
FP6
80%
70%
Other
60%
50%
Industry
40%
Research
30%
Education
20%
10%
0%
returning
new
returning
new
7
Frequency of participation of all types
of entities in FPs.
80%
70%
FP4
60%
FP5
FP6
Entities
50%
40%
30%
20%
10%
0%
1
2
3
4
5
Memberships
6 to 10 11 to 20
>20
8
Structural features of RJV-IST
networks
FP4
FP5
FP6
Nodes
4,879
7,406
3,879
Edges
35,694
73,309
80,437
60
131
7
Giant component (size)
4,618
7,038
3,854
% of total network
94.7
95.1
99.2
Density
0.003
0.0027
0.0107
Stand. Deviation (density)
0.0547
0,0517
0.1029
Characteristic path length
3.39
3.15
2.373
7
7
6
Clustering coefficient
0.854
0.842
0.846
Mean degree
14.69
19.8
41.47
Stand. Deviation (degree)
21.99
38.104
70.903
Normalised degree (avg)
0.3
0.268
1.069
No. of components
Diameter
9
Degree distributions
Frequency of entities
1000
1
FP4
FP5
FP6
Power (FP4)
FP4
FP5
FP6
Power (FP4)
0,1
P(k)
100
0,01
10
y = 160,2x -2,12
R² = 0,957
0,001
0,0001
1
1
10
100
Dergee (k)
1000
10000
1
10
100
Degree (k)
1000
10000
10
7
6
FP5
5
4
3
2
1
0
No of previous links
relative probability of new links
relative probability of new links
Preferential attachment
2
FP6
1,5
1
0,5
0
No of previous links
11
Network resilience
40%
20%
0%
0% 2% 4% 6% 8% 10%
6
4
2
0
0% 2% 4% 6% 8% 10%
f
40%
20%
0%
60%
40%
20%
0%
0% 2% 4% 6% 8% 10%
f
f
Ra ndom nodes
Most connected nodes
Characteristc path length
Characteristic path length
8
60%
80%
0% 2% 4% 6% 8% 10%
f
Ra ndom nodes
Most connected nodes
10
80%
100%
Ra ndom Nodes
Most connected nodes
15
Characteristic path length
60%
100%
Size of giant component
100%
80%
FP6
FP5
Size of giant component
Size of giant component
FP4
12
9
6
3
0
6
5
4
3
2
1
0
0% 2% 4% 6% 8% 10%
0% 2% 4% 6% 8% 10%
f
f
12
Small-world property
FP
4th
5th
6th
Network
Characteristic
path length
Clustering
coefficient
Actual
3.39
0.854
Random
3.164
0.003
Actual
3.15
0.842
Random
2.984
0.003
Actual
2.373
0.846
Random
2.218
0.011
13
Top 1% and top 5% central actors
(based on a composite index of four centrality
measures)
Top 5%
Top 1%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
FP4
FP5
FP6
Education Research Industry Other
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
FP4
FP5
FP6
Education Research Industry Other
14
Conclusions


The networks examined:
 display characteristics typical for complex
networks, such as scale-free distributions and
small-world property
 are highly-connected and robust, gaining in
connectivity through the years
 are dependent on a core of central actors
Educational institutions and research centers have a
more active and prominent role compared to firms
and ‘other’ actors.
15
Suggestions for future research


The analysis of other thematic areas in order
to study regularities or structural differences
due to different knowledge bases, research
organization, industry dynamics and policy
choices related to the specific area.
The conduction of survey work and in-depth
case studies complementary to network
analysis.
16