Nof S.Y. Springer Handbook of Automation

Подождите немного. Документ загружается.

Collaborative Analytics for Astrophysics Explorations 93.5 Sunfall Architecture and Components 1665

mined using a lookup table based on the phase and

redshift of the target. To assist the scheduler in plan-

ning the night’s observations, the Visible Time tool was

implemented (Fig.93.18).

VisibleTimeDisplay

The Visible Timeplot visuallydepicts thenight’s sched-

ule by displayingat a given hour what targets are visible

in the sky. In addition, the plot shows the start times

and the length of the exposure scheduled for each tar-

get. The exposure is color-coded according to the type

of observation, a spectrum,a photometric screen, or a ﬁ-

nal reference. Green bars indicate supernovae that the

project is following, i.e., an interesting object where

a series of spectra have been obtained over time, and

more are desired. Blue indicates a spectroscopic screen-

ing is desired, because the type of the object has not

yet been discovered. A marker on the plot (the zero line

above) indicates the current time, which along with the

visual representation of the schedule, helps the shifter

keep track of events. Feedback from scientists showed

that the visibility chart aided in decision-making, eas-

ing decisions to cancel an exposure or to insert an extra

observation in the schedule.

Post Mortem

Once observation at the telescope is complete, the data

products (images and spectra) from the telescope are

registered with the database and marked with an excla-

mation mark in the SNwarehouse overview page. The

next task is spectral processing and analysis, known as

the post mortem process. The scientist performing post

mortem ﬁlters fortargetswith new data productsthat are

being followed (highlighted in green and blue, depend-

ing on the type of follow-up) and opens up the details

view to evaluate the telescope results (Fig.93.19).

The raw data from the SNIFS spectrograph is com-

plex and requires a signiﬁcant amount of processing to

yield meaning to the scientist. A visual depiction of

the accuracy of the telescope aim and signal strength

provides more information more quickly than tables of

numeric data. The two small images (postage stamps)

on the left-hand side of each observation subwindow

(Fig.93.16) are a custom visualization designed by sci-

entists to indicate whether the telescope is accurately

pointed at the target and whether a good signal was

received by the spectrograph.

The data taken by SNIFS consists of two channels,

blue and red. A channel produces a spectral decomposi-

tion of each sample of a 15×15 pixel image of a 6



×6



square of the sky. Thus, each channel produces an x by

y (position) by λ (wavelength) datacube.Tomakethe

SNwarehouse postage stamps, the datacube is collapsed

(averaged) along the λ-axis, leaving an x by y image.

Since this image averages over many wavelengths, it is

more accurate than a single wavelength slice, roughly

as (N

slices

)

1/2

, where N

slices

is the datacube dimension

in the wavelength direction. The program that performs

Fig. 93.20 A successful spectral observation. The bright

spot in the center represents the supernova, and indicates

that the target was centered by the spectrograph and the

signal was good

Fig. 93.21 A failed observation. No obvious target appears

visible, so no useful spectral data was captured. This could

happen due to a telescope pointing mishap, say the dome

closed due to a humidity spike

a) b)

Fig. 93.22a,b A marginal observation. In (b) abluehalo

surrounds the bright dot in the center, indicating that too

much noise from the supernova’s host galaxy has been

picked up, skewing the spectral results

Part I 93.5

1666 Part I Home, Ofﬁce, and Enterprise Automation

0.3 0.4

spec_06_134_061_002_17_R.fits (T=1000s, z=1.11)

spec_06_134_061_002_17_B.fits (T=1000s, z=1.11)

0.5 0.6 0.7 0.8 0.9 1

Flux

Wavelength λ (×10

Å)

0.8

0.6

0.4

0.2

–0.2

×10

–15

erg Å

SNF20060512-001

0.3 0.4

spec_06_159_063_003_17_R.fits (T=1800s, z=1.10)

spec_06_159_063_003_17_B.fits (T=1800s, z=1.10)

0.5 0.6 0.7

SNF20060512-001

0.8 0.9 1

Flux

Wavelength λ (×10

Å)

×10

–16

erg Å

Fig. 93.24 Spectrum of the same type Ia supernova at 10days af-

ter peak brightness. Note the different proﬁle of the spectrum. To

a domain expert, each of the peaks and troughs has meaning

Fig. 93.23 A spectrum from a very early type Ia super-

nova, 15days before peak brightness. The green and blue

plots indicate the spectral data. The spiky grey lines in the

background depict the background sky spectrum. This su-

pernova is just starting to show the features that deﬁne it as

type Ia



this operation is run as part of the real-time SNIFS

pipeline that is run after each spectrum is read out.

Color-coding and position indicate the accuracy and

signal strength of the received data. If there is a bright

dot in the center of both squares, that is a clear indicator

of a good pointing (Fig.93.20).

A failed observation contains no clear bright cir-

cle in the center of the image Fig.93.21.Amarginal

observation may contain a blue halo, indicating that

too much background noise from the supernova’s host

galaxy is present, and that the spectral data may be

skewed (Fig.93.22).

Supernova scientists, like many domain experts,

demonstrate strong visual pattern-recognition ability

in their ﬁeld of expertise. They can take a single

glance at a picture of a complex spectrum, and in-

stantly determine its type, age in days before or

after peak brightness, and whether it exhibits any un-

usual properties. An early type Ia supernova (captured

well before peak brightness) will display a certain

pattern in its spectral plot (Fig.93.23). A later super-

nova (imaged a few days past peak brightness) will

show other characteristic spectral lines and features

(Fig.93.24). The spectral data are displayed in opti-

mal form in order to facilitate domain experts’ visual

pattern-recognition ability. Spectral data are plotted in

green and blue. The spiky grey lines depict the spec-

trum of the background sky. The broad grey bands

represent areas of atmospheric absorption. Due to the

complexity of the data (thousands of points of ﬂux

versus wavelength for each observation) and the ne-

cessity to make rapid, accurate decisions in order to

maximize the use of limited, expensive telescope time,

visualization provides the most efﬁcient solution to

the problem. The scientist can also access the raw

numeric data by clicking on any of the images or

spectra.

93.6 Conclusions

Sunfall, a collaborative visual analytics system in

operation in a large-scale astrophysics project, has

demonstrated that such systems can have a positive

impact on data-intensive science. In July 2005, at the

commencement of the Sunfall project, the SNfactory

had discovered no supernovae. As of November 2008,

Part I 93.6

Collaborative Analytics for Astrophysics Explorations 93.6 Conclusions 1667

the SNfactory has discovered over 1000 supernovae, of

which some 600 are spectroscopically conﬁrmed and

nearly 400 of which are the critical type Ia SN so useful

for cosmological research. By focusing on improve-

ments in automation, Sunfall has led to more efﬁcient

data processing and signiﬁcant labor savings for project

scientists, enabling them to concentrate on physics re-

search, including supernova discovery and analysis, and

scientiﬁc publications, rather than on repetitive tasks.

See Table 93.2for supernova search data processing im-

provements, and Table 93.3 for details on labor savings

(over ﬁve full-time equivalent staff per year) realized

by the SNfactory project since the development of

Sunfall.

Additionally, the framework architecture, efﬁcient

visual interfaces, and algorithms developed for Sunfall

may be transferable to future large-scale automated

transient alert pipelines such as the Joint Dark Energy

Mission (JDEM) [93.13] or the Large Synoptic Survey

Telescope (LSST) [93.44].

Sunfall search and the Sunfall status monitor, by in-

corporating sophisticated image-processing algorithms

and machine learning capabilities with a usable, highly

interactive visual interface designed to facilitate col-

Table 93.2 SNfactory supernova search data rates before

and after improvements

Oct. 2005 Oct. 2006

(before search (after search

improvements) improvements)

Square degrees of sky 215 725

processed per night

Processing time < 24h < 24h

Supernova selection 60% 75%

efﬁciency

Objects scanned 1094 113

by humans per day

Table 93.3 SNfactory labor rates in full-time equivalent

staff(FTE) by taskbefore andafter Sunfallimplementation

Tasks 2005 (FTE 2007 (FTE

before Sunfall) after Sunfall)

Search 1 0.2

Scanning 3 0.25

Vetting 1 0.15

Scheduling 0.5 Automated

Postmortem 0.5 0.15

Total 6 FTE 0.75 FTE

laborative decision-making, reduced the number of

false-positive supernova candidates by a factor of ten

and reduced scientist scanning workload by 90%.

Sunfall data taking, the visual analytics subsystem

for astrophysicists collaborativelyoperating a largetele-

scope for time-critical supernova observation, demon-

strated the effectiveness of a simplifying visualization,

projecting three-dimensional data to a rectilinear two-

dimensional format, in enabling users to synthesize

large amounts of streaming data and make critical

decisions under time pressure. This conﬁrms previ-

ous studies [93.26, 28, 45] on the value of simplifying

visualizations in interfaces designed for cognitively

overloaded users. Furthermore, we demonstrated that

the cultivation and maintenance of users’ situation

awareness, a concept ﬁrst studied in the ﬁeld of avia-

tion and cockpit management, is crucial in other ﬁelds

where time-critical decisions involving large, dynamic

data sets must be made [93.2,46].

SNfactoryastrophysicists are currentlyusing Sunfall

and the Supernova Warehouse regularly for their sci-

entiﬁc work. We have documented numerous favorable

comments on the interface, and noted in user observa-

tions that less time is spent on routine tasks [93.2,43].

Perhaps the most revelatory critique is that the scientists

have ceased using all other legacy tools that SNware-

house was intended to replace, and they continue to use

the Sunfall framework even though the software devel-

opers have moved on to other projects [93.2,36].

Finally, perhaps one of the most important lessons

to be learned from this case study is the importance

of forming closely integrated, interdisciplinary teams

where members from all ﬁelds are involved in the par-

ticipatory design process from the beginning of the

project. Designing effective interfaces for complex,

time-critical scientiﬁc decision-making requires both

domain expertise and software expertise, and the two

must be closely intertwined for the ultimate success of

the project. Computer scientists and physicists all par-

ticipated in both the scientiﬁc research and the software

development process, which included regular group

meetings and informal code reviews, thus facilitating

the marriage of sophisticated science algorithms with

software engineering best practices. Sunfall is highly

usable software that has received praise from project

scientists and has increased user satisfaction, efﬁciency,

and productivity. Most important of all, Sunfall has fa-

cilitated the production of scientiﬁc data, results, and

publications – the bottom line of science.

Part I 93.6

1668 Part I Home, Ofﬁce, and Enterprise Automation

93.6.1 Future Research Directions

Since many scientists prefer to develop their own soft-

ware, and indeed, ﬁnd writing code a requirement

to accomplish their scientiﬁc work, the scientiﬁc do-

main has characteristics that are very distinct from the

world of commercial software development. In previ-

ous decades, scientists have written their own code,

frequently developing software individually and for

their own use. However, as the complexity and volume

of scientiﬁc data sets continue to explode, process-

ing requirements have been growing too intricate for

single-user development.As a result, teams of computer

scientists and software engineers have been brought in

to implement software packages according to scientists’

speciﬁcations. Nonetheless, due to the nature of scien-

tiﬁc research and the uncertainty of the ﬁnal structure

of science data sets, there have still been pitfalls in

this approach. One challenge is that speciﬁcations are

constantly evolving, and some scientiﬁc needs cannot

be articulated before the scientists are actually working

with the data.

Because of this, tightly coupled communication and

feedback are becoming more and more necessary, and

participatory design can yield substantial beneﬁts. The

scientists we worked with believe that interdisciplinary

teams of software developers will become more com-

mon infuture data-intensivescience projects. There was

agreement that computer scientists should be embed-

ded as full collaborators in science projects, so that my

problems become your problems. At the same time, par-

ticipatory design allows scientists to be an active part of

the design process and gives them a sense of ownership

of the software.

As scientiﬁc projects are being deluged with ex-

ponentially growing quantities of increasingly complex

and dynamic data, there has been a corresponding

acceleration in the rate of change in the scientiﬁc cul-

ture of software development. The escalating pace of

scientiﬁc discovery is being accompanied by new, inter-

disciplinary approaches to scientiﬁc computation, with

a resulting inﬂux of paradigms from relatively young

disciplines such as human–computer interactionand us-

ability engineering.

Although scientists have typically not ranked us-

ability as a major concern in software design, we

believe that is changing. We would like to see fur-

ther case studies of large-scale scientiﬁc projects as

they take usability into account throughout the design

and implementation of their computing frameworks.

We believe study of the human interface to sci-

entiﬁc software will become especially critical for

data-intensive science projects as the gap between

the exponential growth of scientiﬁc data and the

relatively unchanging nature of human cognitive ca-

pacity continues to expand. We call for continued

research into the area of scientist–computer interac-

tion.

References

93.1 G. Aldering, G. Adam, P. Antilogus, P. Astier, R. Ba-

con, S. Bongard, C. Bonnaud, Y. Copin, D. Hardin,

F. Henault, D. Howell, J.-P. Lemonnier, J.-M. Levy,

S. Loken, P. Nugent, R. Pain, A. Pecontal, E. Pecon-

tal, S. Perlmutter, R. Quimby, K. Schahmaneche,

G. Smadja, W. Wood-Vasey: Overview of the nearby

supernova factory, SPIE Symp. Electron. Imaging

(2002)

93.2 C. Aragon, S. Bailey, S. Poon, K. Runge,

R.C. Thomas: Sunfall: A collaborative visual ana-

lytics system for astrophysics, J. Phys. Conf. Ser.

125, 012091 (2008)

93.3 S. Perlmutter, G. Aldering, G. Goldhaber, R.A. Knop,

P. Nugent, P.G. Castro, S. Deustua, S. Fabbro,

A. Goobar, D.E. Groom, I.M. Hook, A.G. Kim,

M.Y. Kirn, J.C. Lee, N.J. Nunes, R. Pain, C.R. Penny-

packer, R. Quimby, C. Lidman, R.S. Ellis, M. Irwin,

R.G. McMahon, P. Ruiz-Lapuente, N. Walton,

B. Schaefer, B.J. Boyle, A.V. Filippenko, T. Math-

eson, A.S. Fruchter, N. Panagia, H.J.M. Newberg,

W.J. Couch: Measurements of omega and lambda

from 42 high-redshift supernovae, Astrophys. J.

517, 565–586 (1999)

93.4 A.G. Riess, A.V. Filippenko, P. Challis, A. Cloc-

chiatti,A.Diercks,P.M.Garnavich,R.L.Gilliland,

C.J. Hogan, S. Jha, R.P. Kirshner, B. Leib-

undgut, M.M. Phillips, D. Reiss, B.P. Schmidt,

R.A. Schommer, R.C. Smith, J. Spyromilio, C. Stubbs,

N.B. Suntzeff, J. Tonry: Observational evidence

from supernovae for an accelerating universe and a

cosmological constant, Astrophys. J. 116,1009–1038

(1998)

93.5 NEAT: Near Earth Asteroid Tracking (Califor-

nia Institute of Technology/JPL, Pasadena 2007),

http://neat.jpl.nasa.gov

93.6 QUEST: Palomar-QUEST Survey (Yale Univ., New

Haven 2007), http://hepwww.physics.yale.edu/

quest/palomar.html

93.7 C. Aragon, D.B. Aragon: A fast contour descriptor

algorithm for supernova image classiﬁcation, SPIE

Part I 93

Collaborative Analytics for Astrophysics Explorations References 1669

Symp. Electron. Imaging: Real-Time Image Process.

(San Jose 2007)

93.8 S. Bailey, C. Aragon, R. Romano, R.C. Thomas,

B.A. Weaver, D. Wong: How to ﬁnd more su-

pernovae with less work: object classiﬁcation

techniques for difference imaging, Astrophys. J.

665, 1246–1253 (2007)

93.9 R. Romano, C. Aragon, C. Ding: Supernova recogni-

tion using support vector machines, 5th Int. Conf.

Mach. Learn. Appl. (ICMLA 2006) (Orlando 2006)

93.10 W.M. Wood-Vasey: Rates and Progenitors of Type Ia

Supernovae. Ph.D. Thesis (University of California,

Berkeley 2004)

93.11 C.T. Zahn, R.Z. Roskies: Fourier descriptors for plane

closed curves, IEEE Trans. Comput. 21(3), 269–281

(1972)

93.12 L. Mazuray, P.J. Rogers: Optical design and engi-

neering, Proc. SPIE 5249, 146–155 (2004)

93.13 NASA: The Joint Dark Energy Mission (NASA, Wash-

ington 2008), http://universe.nasa.gov/program/

probes/jdem.html

93.14 M. Sako, B. Bassett, A. Becker, D. Cinabro, F. De-

Jongh, D. Depoy, M. Sako, B. Bassett, A. Becker,

D. Cinabro, F. DeJongh, D.L. Depoy, B. Dilday,

M. Doi, J.A. Frieman, P.M. Garnavich, C.J. Hogan,

J.Holtzman,S.Jha,R.Kessler,K.Konishi,

H. Lampeitl, J. Marriner, G. Miknaitis, R.C. Nichol,

J.L. Prieto, A.G. Riess, M.W. Richmond, R. Ro-

mani, D.P. Schneider, M. Smith, M. Subba Rao,

N. Takanashi, K. Tokita, K. van der Heyden, N. Ya-

suda, C. Zheng, J. Barentine, H. Brewington,

C. Choi, J. Dembicky, M. Harnavek, Y. Ihara, M. Im,

W. Ketzeback, S.J. Kleinman, J Krzesi

nski, D.C. Long,

E. Malanushenko, V. Malanushenko, R.J. McMil-

lan, T. Morokuma, A. Nitta, K. Pan, G. Saurage,

S.A. Snedden: The Sloan digital sky survey-II su-

pernova survey: search algorithm and follow-up

observations, Astron. J. 135, 348–373 (2008)

93.15 P.Astier,J.Guy,N.Regnault,R.Pain,E.Aubourg,

D. Balam, S. Basa, R.G. Carlberg, S. Fabbro,

D. Fouchez, I.M. Hook, D.A. Howell, H. Lafoux,

J.D. Neill, N. Palanque-Delabrouille, K. Perrett,

C.J. Pritchet, J. Rich, M. Sullivan, R. Taillet,

G. Aldering, P. Antilogus, V. Arsenijevic, C. Bal-

land, S. Baumont, J. Bronder, H. Courtois, R.S. Ellis,

M. Filiol, A.C. Gonçalves, A. Goobar, D. Guide,

D. Hardin, V. Lusset, C. Lidman, R. McMahon,

M. Mouchet, A. Mourao, S. Perlmutter, P. Ripoche,

C. Tao, N. Walton: SuperNova legacy survey: mea-

surement of Ω

, Ω

and w from the ﬁrst year data

set, Astron. Astrophys. 447,31–48(2006)

93.16 J. Raddick: Sloan Digital Sky Survey (Johns Hopkins

Univ., Baltimore 2007), http://www.sdss.org

93.17 SNLS: SuperNova Legacy Survey (2007),

http://www.cfht.hawaii.edu/SNLS/

93.18 J. Thorstensen: SkyCalc GUI Manual (Dartmouth

College, Hanover 2008),

http://mdm.kpno.noao.edu/Manuals/doc/

guimanual.html

93.19 J. Thorstensen: SkyCalc User’s Manual (Dartmouth

College, Hanover 2008),

http://zimmer.csufresno.edu/˜fringwal/skycal.pdf

93.20 F. Cavicchio: Astroart 4.0–96 bit image processing

(MSB Software, Ravenna 2008),

http://www.msb-astroart.com/

93.21 L. Christensen: ESA/ESO/NASA Photoshop FITS Liber-

ator (ESO, Munich 2008),

http://www.spacetelescope.org/projects/

ﬁts_liberator/index.html

93.22 A. Gauthier: Virtual Astronomy Multimedia Project

(University of Arizona, Tempe 2008),

http://www.virtualastronomy.org/index.php

93.23 W. Joye, E. Mandel: New features of SAOImage DS9,

Astronomical Data Analysis Software and Systems

XII (2003)

93.24 T. Boroson: NOAO Image Reduction and Analysis

Facility (IRAF, National Optical Astronomy Observa-

tories, Tucson 2008), http://iraf.noao.edu/iraf/web

93.25 Y. Li, C. Fu, A. Hanson: Scalable WIM: Effective

exploration in large-scale astrophysical environ-

ments, IEEE Trans. Vis. Comput. Graph. 12(5),

1005–1011 (2006)

93.26 L. Chittaro, F. Zuliani, E. Carchietti: Mobile devices

in emergency medical services: user evaluation

of a PDA-based interface for ambulance run re-

porting. In: Mobile Response, Lecture Notes in

Computer Science, Vol.4458 (Springer, Berlin, Hei-

delberg 2007) pp. 19–28

93.27 S. Eick, M. Eick, J. Fugitt, B. Horst, M. Khailo,

R. Lankenau: Thin client visualizatio, IEEE Symp.

Vis. Anal. Sci. Technol. (VAST 2007) (Sacramento

2007)

93.28 S. Kim, Y. Jang, A. Mellema, D. Ebert, T. Collins:

Visual analytics on mobile devices for emergency

response, IEEE Symp. Vis. Anal. Sci. Technol. (VAST

2007) (Sacramento 2007)

93.29 A. Pattath, B. Bue, Y. Jang, D. Ebert, X. Zhong,

A. Ault, E. Coyle: Interactive visualization and anal-

ysis of network and sensor data on mobile devices,

IEEE Symp. Vis. Anal. Sci. Technol. (VAST 2007) (Bal-

timore 2006)

93.30 Purdue: Synthetic Environment for Analysis and

Simulation (Purdue Univ., West Lafayette 2008),

http://www.mgmt.purdue.edu/centers/perc/html

93.31 T. Kapler, W. Wright: Geotime information visual-

ization, IEEE Symp. Inf. Vis. (2004)

93.32 C. Pan, P. Mitra: FemaRepViz: Automatic extraction

and geo-temporal visualization of FEMA national

situation updates, IEEE Symp. Vis. Anal. Sci. Tech-

nol. (VAST 2007) (Sacramento 2007)

93.33 RSOE: RSOE HAVARIA AlertMap (Emergency and Dis-

aster Information Services (EDIS), (Budapest 2008),

http://hisz.rsoe.hu/alertmap/

woalert.php?lang=eng

Part I 93

1670 Part I Home, Ofﬁce, and Enterprise Automation

93.34 D. Tesone, J. Goodall: Balancing interactive data

management of massive data with situational

awareness through smart aggregation, Proc. IEEE

Symp. Vis. Anal. Sci. Technol. (VAST 2007) (Sacra-

mento 2007)

93.35 M.R. Endsley, D.J. Garland: Situation Awareness

Analysis and Measurement (Lawrence Erlbaum,

New York 2000)

93.36 C. Aragon, S. Poon: The impact of usability on

supernova discovery, Workshop on Increasing the

Impact of Usability Work in Software Development,

CHI 2007: ACM Conf. Hum. Factors Comput. Syst. (San

Jose 2007)

93.37 J.J. Thomas, K.A. Cook: Illuminating the Path: The

Research and Development Agenda for Visual Ana-

lytics (National Visualization and Analytics Center,

Richland 2005)

93.38 R. Chellappa, R. Bagdazian: Fourier coding of im-

age boundaries, IEEE Trans. Pattern Anal. Mach.

Intell. 6, 102–105 (1984)

93.39 E. Persoon, K. Fu: Shape discrimination using

Fourier descriptors, IEEE Trans. Syst. Man Cybern.

7, 170–179 (1977)

93.40 R. Mak, J. Walton, L. Keely, D. Heher, L. Chan:

Reliable service-oriented architecture for NASA’s

Mars Exploration Rover Mission, IEEE Aerosp. Conf.

(2005)

93.41 R. Israel: Mercator’s Projection (Univ. British

Columbia, Vancouver 2008), http://www.math.

ubc.ca/˜israel/m103/mercator/mercator.html

93.42 K. Arnold: Screen Updating and Cursor Movement

Optimization: A Library Package (University of Cal-

ifornia, Berkeley 1977)

93.43 S. Poon, R.C. Thomas, C. Aragon, B. Lee: Context-

linked virtual assistants for distributed teams: an

astrophysics case study, ACM Conf. Comput. Sup-

ported Coop. Work (ACM CSCW 2008) (San Diego

2008)

93.44 Z. Ivezic, J.A. Tyson, R. Allsman, J. Andrew, R. Angel,

for the LSST collaboration: LSST: from Science drivers

to reference design and anticipated data products,

arXiv:0805.2366v1 [astro-ph] (2008)

93.45 C. Aragon, M. Hearst: Improving aviation safety

with information visualization: a ﬂight simulation

study, CHI 2005: ACM Conf. Hum. Factors Comput.

Syst. (Portland 2005)

93.46 C. Aragon, S. Poon, G. Aldering, R.C. Thomas,

R. Quimby: Using visual analytics to maintain situ-

ational awareness in astrophysics, IEEE Symp. Vis.

Anal. Sci. Technol. (VAST 2008) (Columbus 2008)

Part I 93

1671

Appendix

Part J

Part J Appendix

94 Automation Statistics

Juan D. Velásquez, West Lafayette, USA

Xin W. Chen, West Lafayette, USA

SangWonYoon,WestLafayette,USA

Hoo Sang Ko, West Lafayette, USA

1672

Appendix. Part J The concluding part of this Handbook contains ﬁgures and tables with statistical information

and summaries about automation applications and impacts in three main areas: industrial automation; service

automation; and ﬁnancial and e-Commerce automation. A rich list of associations and of periodical publications

around the world that focus on automation in its variety of related ﬁelds is also included for the beneﬁt of readers

worldwide.

1673

Automation S

94. Automation Statistics

Juan D. Velásquez, Xin W. Chen, Sang Won Yoon, Hoo Sang Ko

This chapter contains automation information col-

lected from many different sources around the

world, among them the US Census Bureau, the In-

ternational Federation of Automatic Control (IFAC),

the American Automatic Control Council (AACC),

the American Bankers Association, World Robotics,

and many more. Section 94.1 introduces automa-

tion statistical data according to its application

domain, such as: ﬁnancial and e-Commerce, in-

dustrial, healthcare, and the service industries. In

Sect. 94.2 a list of worldwide automation, control,

and robotics associations is provided. Section 94.3

provides the reader with a list of automation labs

around the world and lastly, in Sect. 94.4 alistof

automation related journals and publications is

presented.

The statistics, organizations, and journals

included in this chapter open a window into the

current and emerging state of automation. While

attempting to be comprehensive, because of the

broadness of automation, the authors recognize

that there may be additional relevant items, but

none were omitted intentionally. The scope of

94.1 Automation Statistics............................1674

94.1.1 Financial

and e-Commerce Automation........1674

94.1.2 Industrial Automation ..................1677

94.1.3 Automation in Service

and the Healthcare Industry ..........1680

94.1.4 Service Automation ......................1682

94.2 Automation Associations .......................1685

94.3 Automation Laboratories

Around the World .................................1693

94.4 Automation Journals

from Around the World .........................1696

94.4.1 Automation-Related Journals ........1698

the information is meant to highlight and ex-

pose the broad span, applications, concerns

and beneﬁts of automation, and clearly, can-

not completely include all areas of automation

inﬂuence. Beyond this chapter, other chapters

in this handbook provide additional statis-

tical data directly related to their speciﬁc

topic.

Automation continues toenable individuals, companies,

and governments to innovate and compete in a fast-

changing world. Companies of all sizes throughout the

world in virtually all industries have embraced automa-

tion as one of the most powerful strategic advantages to

stay competitive. For instance, in robotic automation it

is expected that over 1million robots will be installed

worldwide by 2011, a signiﬁcant gain from the 400000

that were installed in the early 1990s. Japan, the USA,

and Germany will continue to supply the majority of

robots, while automotive parts and motor vehicles will

continue to demand the greatest number of robots by

industry.

It has not only been companies that are taking ad-

vantage of automation; individuals have also embraced

automation as a way to enhance their everyday lives.

It is expected that, by 2012, consumers’ most used

payment form will be electronic transfers, followed by

credit cards. More than 50% of all travel reservations

made in the USA are now done by consumers directly

via the Internet rather than by travel agencies. Like-

wise, the number of online buyers continues to climb,

with over 50% of Americans having bought at least one

product online in 2007. It is projected that in the USA

alone e-Commerce will grow from US$282.7billion in

2007 to US$ 658.4 billion by 2010, almost a threefold

increase in 3years.

What does the future hold for automation? Au-

tomation’s inﬂuence on microsystems, nanosystems,

and systems-of-systems will explode, as will bioin-

Part J 94

1674 Part J Appendix

spired automationand bioinspiredcollaborative control.

There are many grand challenges as identiﬁed by

the US National Academy of Engineering for which

automation is essential (i. e., make solar energy eco-

nomical, advance personalized learning, engineer better

medicines). For instance, virtual reality will signiﬁ-

cantly inﬂuence healthcare, engineering, entertainment,

and education and training as its worldwide market

grows from US$2billion in 2002 to over US$ 8 billion

in 2010.

94.1 Automation Statistics

94.1.1 Financial

and e-Commerce Automation

328.7

329.7

2007

146.3

282.4

136.4

2000

29.4

57.4

2001

38.1

73.4

35.3

2002

94.8

46.8

2003

57.3

117.3

2004

83.3

159.6

76.3

2005

95.7

189

93.3

2006

117.4

232.1

114.6

a) Revenue (×10

US$)

Year

350

300

250

200

150

100

Total e-commerce service

Total e-commerce retail

Total B2C e-commerce

b) Revenue (×10

US$)

Year

700

600

500

400

300

200

100

Total e-commerce service

Total e-commerce retail

Total B2C e-commerce

282.7

2007

146.3

136.4

166.8

156

191.7

171.9

226.4

211.8

271.7

263.5

322.7

2008

363.6

2009

438.2

2010

535.2

2011

658.4

2012

Fig. 94.1a,b Business-to-consumer

e-Commerce

(a) revenue by service

and retail sectors for 2000–2007 in

US$.

(b) Projected revenue by ser-

vice and retail sectors for 2007–2012

in US$ (source: US Department of

Commerce, US Census Bureau)

Part J 94.1