Presentation is loading. Please wait.

Presentation is loading. Please wait.

Miguel C ASQUILHO Assist. Professor (retired) DEQDepartment of Chemical Eng. ing CERENA“Centro de Recursos Naturais e Ambiente” (Centre for Natural Resources.

Similar presentations


Presentation on theme: "Miguel C ASQUILHO Assist. Professor (retired) DEQDepartment of Chemical Eng. ing CERENA“Centro de Recursos Naturais e Ambiente” (Centre for Natural Resources."— Presentation transcript:

1 Miguel C ASQUILHO Assist. Professor (retired) DEQDepartment of Chemical Eng. ing CERENA“Centro de Recursos Naturais e Ambiente” (Centre for Natural Resources and the Environment) ISTInstituto Superior Técnico ULisboaUniversity of Lisbon http://web.tecnico.ulisboa.pt/mcasquilho/ In Industry and in Academia simultaneously: the Internet connection IST — Lisboa, 24 April 2015 24-Apr-2015 In Industry and in Academia 1 / 31

2 A BSTRACT  In our technological era, the Internet permits being in Industry and in Academia simultaneously through computing over the Web (CW). It has been overlooked by both, but can provide the link between them. CW = the user (in a website) supplies his data, executes a program, and gets his results. (No software installation.) Based on cases that I use in teaching (via the Web), we’ll follow these points (on scientific CW): 1. Antecedents; 2. Examples; 3. Producer-consumer communication (of CW); and 4. Conclusions — the Web provides an Industry-Academia link. In Industry and in Academia simultaneously: the Internet connection 24-Apr-2015 In Industry and in Academia 2 / 31

3  Cogito, ergo sum. 1596-03-31 1650-02-11  In Interrete non existo, ergo non sum. 24-Apr-2015 Existere: to stand (Pt: estar) In Industry and in Academia 3 / 31

4 1.Antecedents 2.Examples 3.Producer-consumer communication 4.Conclusions In Industry and in Academia simultaneously: the Internet connection Manuel Amado (1938–) Porta da estação 1986 (Station door) 24-Apr-2015 In Industry and in Academia 4 / 31

5  Scientific computing (SC) is arduously taken to Industry…  …but Industry often does not take advantage from SC: Lack of structure — no “surplus” technical personnel for it Lack of stimulus — no disposition; fear We don’t use those things * that you taught us… (sic) * One of my former students, about “Operational Research”. 1. 2. 3. 4. Antecedents 24-Apr-2015 In Industry and in Academia 5 / 31

6  We talk to: Industry, but also other academic institutions.  Yet:  The Web * itself has not paid enough attention to SC (either with free or paid software).  Among countless webpages, there are: manymany (didactic) for consultation, study fewfew for dynamic use, i.e.: data  results  Scarce applicable material is found on Mathematics, Engineering, industrial applications  So, what about solving problems (industrial, academic) ? * http://en.wikipedia.org/wiki/Internet#Services (about: Web  Internet)http://en.wikipedia.org/wiki/Internet#Services 1. 2. 3. 4. Antecedents 24-Apr-2015 In Industry and in Academia 6 / 31

7 Software that exists in our (any) organization (A parenthesis)  One of us (technical staff, professor, student, etc.) can use the software (Matlab  ), to which he can access remotely,to which he can access remotely to solve this simple problem (via Monte Carlo simulation): What is the total length of these 2 (similar) bars, each with ~50 cm ? (Perhaps 1 m !) Suppose uniform distributions in 49–51 cm. On the Web: sum of uniforms  You also can do it !sum of uniforms 24-Apr-2015 In Industry and in Academia 7 / 31

8 Many Many webpages on science and… Wolfram Mathworld, by Eric Weisstein http://mathworld.wolfram.com/ SpringerEncyclopaedia of Mathematics http://eom.springer.de/ Wikipedia (part) http://en.wikipedia.org/ NIST/SEMATECH e-Handbook of Statistical Methods (Statistical Eng. ing Division  Information Technology Laboratory  Nat’l Inst. of Standards and Technology, USA) http://www.itl.nist.gov/div898/handbook/ Etc., etc. 1. 2. 3. 4. Antecedents 24-Apr-2015 In Industry and in Academia 8 / 31

9 few …few webpages on scientific computing (?) Academic webpages — very useful didactic material  J. E. Beasley, Brunel Univ. London — but no calculations J. E. Beasley http://people.brunel.ac.uk/~mastjjb/jeb/jeb.html  Robert Vanderbei, Princeton U. — but really no calc. (Java !) Robert Vanderbei http://www.princeton.edu/~rvdb/JAVA7/pivot/simple.htmlhttp://www.princeton.edu/~rvdb/JAVA7/pivot/simple.html   H. Arsham, Univ. of Baltimore — with calc., awkward H. Arsham http://home.ubalt.edu/ntsbarsh/Business-stat/otherapplets/MultiCorr.htmhttp://home.ubalt.edu/ntsbarsh/Business-stat/otherapplets/MultiCorr.htm   Michael Trick, Princeton U. — but no calculations Michael Trick 1. 2. 3. 4. Antecedents 24-Apr-2015 In Industry and in Academia 9 / 31

10 few …few webpages on scientific computing  John W. Chinneck, Carleton Univ. — example without data John W. Chinneck http://optlab-server.sce.carleton.ca/POAnimations2007/DijkstrasAlgo.html http://optlab-server.sce.carleton.ca/POAnimations2007/DijkstrasAlgo.html   Attractor — in WebMathematica (needs “our” Java * !) http://atractor.pt/webM/exemplos/poliedros.htm  Victor M. Ponce, San Diego State Univ. — exception ! Victor M. Ponce (In Fortran) ~without default dataFortran http://ponce.sdsu.edu/onlinewaterbalance.php  “scientific computing on the internet” “scientific computing on the internet” ~6 results (sources of software, university courses) 1. 2. 3. 4. Antecedents 24-Apr-2015 * Beware since v. 8 ! In Industry and in Academia 10 / 31

11 F ROM MY EXPERIENCE  Computer Science engineers (informatics engineers), generally: Are not enthusiastic about (this) Scientific Computing Are not inclined to Mathematics  The other engineers, generally: Leave Information Technology (Informatics) to the former Fear (justifiably…) the Internet environment 1. 2. 3. 4. Antecedents 24-Apr-2015 In Industry and in Academia 11 / 31

12  In my academic work, “Computing”  “Operational Research”  “Quality Control” I have adopted SC over the Web, since 1998.  My 1. st example: Area of a triangle by Heron’s * formula (year AD 60) Program in Fortran (5 lines) (a, b, c, sides; s, semiperimeter) About 1 year to put it on the Internet http://web.tecnico.ulisboa.pt/mcasquilho/compute/misc/F-triang.html * Heron of Alexandria (AD 10 ?–75 ?)Heron of Alexandria 1. 2. 3. 4. Antecedents 24-Apr-2015 In Industry and in Academia 12 / 31

13 1.Antecedents 2.Examples 3.Producer-consumer communication 4.Conclusions Carlos Botelho (1899-1982) Ramalhete de Lisboa 1935 (Lisbon bouquet) 24-Apr-2015 In Industry and in Academia simultaneously: the Internet connection In Industry and in Academia 13 / 31

14 Chi-square Try = 3, = 50 (nearly Gaussian) http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/Fx-chisquare.phpChi-square  Try = 3, = 50 (nearly Gaussian) http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/Fx-chisquare.php http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/Fx-chisquare.php 1. 2. 3. 4. Examples (Images are hyperlinks) 24-Apr-2015 In Industry and in Academia 14 / 31

15 Fraction defective (Quality Control) Try  low = 10, 1. http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/Fx-fracdefective.phpFraction defective (Quality Control)  Try  low = 10, 1. http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/Fx-fracdefective.php http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/Fx-fracdefective.php 1. 2. 3. 4. Examples 24-Apr-2015 In Industry and in Academia 15 / 31

16 Convergence to non-central t http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/F-tncConverg.phpConvergence to non-central t — Results: 30 000, 10 6 (= 1e6) trials http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/F-tncConverg.php http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/F-tncConverg.php 1. 2. 3. 4. Examples 24-Apr-2015 In Industry and in Academia 16 / 31

17 Packet filling, as tubes (FIFO) http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/Fx-3tubefill.phpPacket filling, as tubes (FIFO)  Result FIFO. Try: none, FIFO. http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/Fx-3tubefill.php http://web.tecnico.ulisboa.pt/mcasquilho/compute/qc/Fx-3tubefill.php 1. 2. 3. 4. Examples 24-Apr-2015 In Industry and in Academia 17 / 31

18 EOQ (economic order q.) with random demand http://web.tecnico.ulisboa.pt/mcasquilho/compute/or/Fx-inventoryRand.phpEOQ (economic order q.) with random demand  Try c p = 2000, 150. http://web.tecnico.ulisboa.pt/mcasquilho/compute/or/Fx-inventoryRand.php http://web.tecnico.ulisboa.pt/mcasquilho/compute/or/Fx-inventoryRand.php 1. 2. 3. 4. Examples 24-Apr-2015 In Industry and in Academia 18 / 31

19 Catenary with concentrated loads * http://web.tecnico.ulisboa.pt/mcasquilho/compute/com/Fx-catenaryLoads.phpCatenary with concentrated loads *  Try 3. rd load -1, -1.9. http://web.tecnico.ulisboa.pt/mcasquilho/compute/com/Fx-catenaryLoads.php http://web.tecnico.ulisboa.pt/mcasquilho/compute/com/Fx-catenaryLoads.php * Problem and resolution: colleague Prof. A. Pinto da Costa, DECivil, IST 1. 2. 3. 4. Examples 24-Apr-2015 In Industry and in Academia 19 / 31

20 1. 2. 3. 4. Examples Semicontinuous cooled reactor http://web.tecnico.ulisboa.pt/mcasquilho/compute/com/Fx-scoolreactor.phpSemicontinuous cooled reactor  Try T = 298, 348 K (25, 75  C). http://web.tecnico.ulisboa.pt/mcasquilho/compute/com/Fx-scoolreactor.php http://web.tecnico.ulisboa.pt/mcasquilho/compute/com/Fx-scoolreactor.php 24-Apr-2015 In Industry and in Academia 20 / 31

21 Minimum distance “within an angle” http://web.tecnico.ulisboa.pt/mcasquilho/compute/com/Fx-angdistImg.phpMinimum distance “within an angle”  Try  = 50 , 5  http://web.tecnico.ulisboa.pt/mcasquilho/compute/com/Fx-angdistImg.php http://web.tecnico.ulisboa.pt/mcasquilho/compute/com/Fx-angdistImg.php 1. 2. 3. 4. Examples (Casquilho & Buescu [2011]) 24-Apr-2015 In Industry and in Academia 21 / 31

22 Distance to points in a circle http://web.tecnico.ulisboa.pt/~mcasquilho/compute/or/Fx-distInCircle.phpDistance to points in a circle  Try x 0 = 3, 1, 0.5, 0 (R = 1) http://web.tecnico.ulisboa.pt/~mcasquilho/compute/or/Fx-distInCircle.php http://web.tecnico.ulisboa.pt/~mcasquilho/compute/or/Fx-distInCircle.php 1. 2. 3. 4. Examples 24-Apr-2015 In Industry and in Academia 22 / 31

23 1.Antecedents 2.Examples 3.Producer-consumer communication 4.Conclusions Almada Negreiros (1893–1970) --- 1928 (No title) 24-Apr-2015 In Industry and in Academia simultaneously: the Internet connection In Industry and in Academia 23 / 31

24 Paraboloid of revolution http://www.gsd.inesc-id.pt/~jcg-test/compute/Fx-parabRemExperts.phpParaboloid of revolution  Compute: here at Técnico http://www.gsd.inesc-id.pt/~jcg-test/compute/Fx-parabRemExperts.php http://www.gsd.inesc-id.pt/~jcg-test/compute/Fx-parabRemExperts.phphere at Técnico http://www.gsd.inesc-id.pt/~jcg-test/compute/Fx-parabRemExperts.php My webpage at Inesc * …My webpage at Inesc * … …does the computing at Técnico…does the computing at Técnico 1. 2. 3. 4. Producer-consumer communication 24-Apr-2015 * Inesc-ID account courtesy of: colleague Prof. João C. Garcia, Inesc-ID In Industry and in Academia 24 / 31

25 For a company, the computing can be: Started on its website; Executed on remote servers (Universities ?); and Presented on the company’s website. The user does not even notice the circuit. 1. 2. 3. 4. Producer-consumer communication 24-Apr-2015 In Industry and in Academia 25 / 31

26 1.Antecedents 2.Examples 3.Producer-consumer communication 4.Conclusions António Dacosta (1914–1991) A festa 1942 (The party) 24-Apr-2015 In Industry and in Academia simultaneously: the Internet connection In Industry and in Academia 26 / 31

27  Scientific computing over the Web has been overlooked can provide easy links — e.g., Industry-Academia can be difficult to start, perhaps justifying its scarcity.  Working on the Web avoids platform incompatibilities (Windows, Mac, Linux) obviates limitations of the user’s terminal (PC, phone) avoids software installation (uninstallation)  Modern (scientific) languages (Mathematica, Matlab, etc.) favour productivity, but appear obsolete (!), as they aren’t (yet) Web-friendly (licences, permissions). 1. 2. 3. 4. Conclusions… 24-Apr-2015 In Industry and in Academia 27 / 31

28  Scientific computing over the Web benefits from sharing — in team work visibility — attracting connections (  business) network computing — “grid”, “cloud”, parallel easy access, just using a browser  Problems: welcome ! 1. 2. 3. 4. …Conclusions 24-Apr-2015 In Industry and in Academia 28 / 31

29  CERENA: “Centro de Recursos Naturais e Ambiente” (Centre for Natural Resources and the Environment), Instituto Superior Técnico (IST), ULisboa (University of Lisbon)  DEQ: Department of Chemical Engineering, IST, UL  CIIST: “Centre of Informatics of IST”, IST, UL  Inesc-ID: “Instituto de Engenharia de Sistemas e Computadores, Investigação e Desenvolvimento”, Inesc Group, Lisbon  Profs. (DEIC & Inesc-ID) João C. Garcia, (DECivil) A. Pinto da Costa, M. Azevedo Coutinho (  Ponce), (DMat, FacC, UL) Jorge Buescu, (DEstClass, FacLetras, UL) Arnaldo Espírito Santo Keywords Scientific computing; Web; Internet; industry, academia Acknowledgements 24-Apr-2015 In Industry and in Academia 29 / 31

30 Casquilho, M. [2012] “Computação científica na Internet, um campo pouco explorado”, Universidade Aberta, Porto Salvo (Oeiras, Portugal), 03 Feb. Casquilho, M., J. Buescu [2011] "A minimum distance: arithmetic and harmonic means in a geometric dispute", International Journal of Mathematical Education in Science and Technology, 142(3), 399–405 (doi:10.1080/0020739X.2010.526253).10.1080/0020739X.2010.526253 Casquilho, M., M. Cunha [2015] “Applied scientific computing over the Web with remote servers: sampling with and without replacement”, CISTI'2015, 10.ª Conf. Ibérica de Sistemas e Tecnologias de Informação (10. th Iberian Conf. on Information Systems and Technologies), Aveiro (Portugal), to be held June 2015.CISTI'2015 Ferreira, M., M. Casquilho [2013] “Scientific computing over the Internet: an example in Geometry”, WorldCIST'13, World Conference on Information Systems and Technologies, 27 Mar., Olhão (Portugal).WorldCIST'13 Franco, B., M. Casquilho [2011] “A Web application for scientific computing: combining several tools and languages to solve a statistical problem”, CISTI'2011, 6.ª Conferência Ibérica de Sistemas e Tecnologias de Informação, Chaves (Portugal).CISTI'2011 Hughes, Alan [2006] “University-industry linkages and UK science and innovation policy”, Working Paper No. 326, Centre for Business Research, University of Cambridge, Cambridge (UK) [“Internet” or “the Internet” ?] in http://microsites.oii.ox.ac.uk/ics2011/ (2011-Sep).http://microsites.oii.ox.ac.uk/ics2011/ Bibliography 24-Apr-2015 In Industry and in Academia 30 / 31

31 Miguel C ASQUILHO Assist. Professor (retired) DEQDepartment of Chemical Eng. ing CERENA“Centro de Recursos Naturais e Ambiente” (Centre for Natural Resources and the Environment) ISTInstituto Superior Técnico ULisboaUniversity of Lisbon http://web.tecnico.ulisboa.pt/mcasquilho/ In Industry and in Academia simultaneously: the Internet connection IST — Lisboa, 24 April 2015 24-Apr-2015 In Industry and in Academia 31 / 31


Download ppt "Miguel C ASQUILHO Assist. Professor (retired) DEQDepartment of Chemical Eng. ing CERENA“Centro de Recursos Naturais e Ambiente” (Centre for Natural Resources."

Similar presentations


Ads by Google