Relations Between Science &
Technology
April 10, 2007
What have we learned?
• So far, no answers but three questions.
– Why don’t those outside the organizations
supply better ideas?
– Why aren’t those inside the organization
consulted more?
– How do the ‘internal consultants’ acquire
their information?
Science and Technology
Verbal
Info
Verbal
Info
Verbal
Info
Scientist
Scientist
Reverse Engineering
Physically
Encoded
Verbal
Info
Info
Engineer
Verbal
Info
Engineer
Verbal
Info
Science, Technology and Market
Physics to Electronics (WW II)
Nuclear Physics to Nuclear Eng. (‘50s & ‘60s)
Biology to Bio Engineering (’90s & ’00s)?
Migration
Science
Technology
Pull
Science
Normal
Science
Technology
Blockage
Market
Pull
Technology
Push
Market
Pull
Technology
Market
Internal Problems
• There are boundaries inside the
organization that can cause as much
difficulty (sometimes more) as the
external boundary:
– Product Development Ù Marketing.
– Product Development Ù Manufacturing
Engineering.
– Etc.
Interfunctional Communication
Product Development
Manufacturing
Engineering
An Experiment
• Recorded CAD functions used by gas
turbine designers.
– Sampled on random days about once per
week for several months.
• Outcome Measure:
– Number of Engineering Changes (ECs)
processed following transfer of design to
manufacturing.
Computer Aided Design and Inter-functional
Communication
• CAD can be implemented on at least
three levels:
– An investment in physical capital.
• An automation of the traditional drafting
process.
– As an investment in human capital.
• Extending the ability of the designer.
– As an investment in social capital.
• As a means for communication and other
exchange.
An Experiment
• Instrumented the terminals of gas
turbine blade designers.
– On randomly chosen days, recorded all of
the functions used that day.
• Functions categorized according to categories
listed on preceding slide.
– Output measure
• Number of Engineering Changes that had to
be processed following transfer to
Manufacturing.
Quality of Transfer
(Fewer ECs)
Performance in Transferring Designs to Manufacturing as
a Function of CAD System Use for Communication
High
p < 0.02
Low
0
0.2
0.4
0.6
0.8
1
1.2
Mean Number of Communications per Day
(Through CAD System)
1.4
Using a Common Reference to Reduce
Ambiguity in Communication
Product Development
Manufacturing Engineering
Figure by MIT OCW.
Partial Layout of the BMW Forschung und
Ingenieurung Zentrum
Product
Development
Prototype
Manufacturing
Engineering
Movement of People
• More may happen as a result of a
transfer than is immediately obvious.
The Effect of Transfers
A
N1
B
Continuing Relations
A
B
Potential Contacts
N1
More Continuing Relations
A
B
Potential Contacts
N1
N2
Referrals
A
B
Referrals
N1
N2
A Typical Technical Communication
Network
Figure by MIT OCW.
Communication Network in a Small
Laboratory
12
14
9
13
7
11
6
20
22
2
1
8
21
5
10
19
24
25
35
18
27
30
4
16
23
34
17
3
33
15
29
26
28
36
31
32
Figure by MIT OCW.
High Communicators Compared with
Colleagues in Readership of Refereed Journals
p < 0.001
Laboratory 'A'
Laboratory 'G' p < 0.02
Laboratory 'E'
Laboratory 'L'
Laboratory 'M'
Laboratory 'H'
Laboratory 'S'
p < 0.01
p < 0.05
p < 0.05
p < 0.01
p < 0.01
p < 0.01
Laboratory 'T'
p < 0.01
Laboratory 'U'
Laboratory 'V'
0
High Communicators
All Other Staff
N.S.
5
10
15
Mean Journal Readership
20
High Communicators Compared with Colleagues in Terms
of Regular Informal Contact Outside of the Organization
p < 0.05
Laboratory 'A'
p < 0.05
Laboratory 'G'
p < 0.001
Laboratory 'E'
p < 0.05
Laboratory 'L'
p < 0.01
Laboratory 'M'
Laboratory 'H'
p < 0.01
p < 0.01
Laboratory 'F'
0
High Communicators
All Other Staff
1
2
3
4
5
Mean Number of Contacts
6
The Gatekeeper as a Link to Outside
Technology
Outside
Experts
X X X
X X
Gatekeeper
Literature