Engineering coordination [Elektronische Ressource] : a methodology for the coordination of planning systems / von René Schumann

Sujets

• Informatik

Informations

• Publié par: goethe_universitat_frankfurt_am_main
• Publié le: 01 janvier 2010
• Nombre de lectures: 22
• Langue: English
• Poids de l'ouvrage: 4 Mo

Extrait

Engineering Coordination
A Methodology for the Coordination of Planning Systems
Dissertation
zur Erlangung des Doktorgrades
der Naturwissenschaften
vorgelegt beim Fachbereich Informatik und Mathematik
der Johann Wolfgang Goethe { Universit at
in Frankfurt am Main
von
Rene Schumann
aus Delmenhorst
Frankfurt (2010)
(D 30)vom Fachbereich Informatik und Mathematik der
Johann Wolfgang Goethe { Universit at als Dissertation angenommen
Dekan: Prof. Dr. Tobias Weth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Gutachter: Prof. Dr. Ingo J. Timm, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Prof. Dr. Jurgen Sauer, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Prof. Dr. Ulrich Schwanecke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Datum der Disputation: 06.10.2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Acknowledgement
The work this thesis is based on, has been done over several years and at three
di erent locations. During my appointment at the OFFIS Institute for Informa-
tion Technology in Oldenburg, while I was employed at the Goethe University in
Frankfurt am Main, and during my research stays at the Arti cial Intelligence
group at the Rey Juan Carlos University in M ostoles (Spain).
The issue of coordination of autonomous planning systems has been at the center
of my research for more than six years now. It started when I was a student at
the Carl von Ossietzky University Oldenburg. This project took longer than I
expected and I moved to locations I never thought I would visit when I started it.
Consequently, over the course of time I have to thank a number of people I have
meet during this time at the various locations.
First of all, I would like to thank my main supervisor Prof. Dr. Ingo J. Timm
(Goethe University Frankfurt am Main). I met him in 2005 at a conference and
moved to Frankfurt in 2007 to become part of his group. I would like to thank him
for the opportunity to move to Frankfurt. It was not only a geographical change,
but also o ered me a complete new perspective on research. I am very grateful
for all the valuable advice and encouragement that he still gives me.
Furthermore, I would like thank apl. Prof. Dr. Jur gen Sauer (Carl von Ossiet-
zky University Oldenburg). He supervised my diploma thesis in 2004 and to this
day still gives me valuable advice. He encouraged me to start with my PhD and
support me continuously even after I left Oldenburg.
I also want to thank Prof. Dr. Ulrich Schwanecke (Hochschule RheinMain
University of Applied Sciences) for the valuable advice he gave me since I joined
the Sensyble graduate school.
I am deeply grateful for the hospitality and valuable discussions of Prof. Dr.
Sascha Ossowski and his entire group. I spent 10 days in June 2009 and three
month (Feb.{Apr.) in 2010 as a visiting researcher in his group. Large parts of
this thesis have been written there.
I want to express my gratitude to Dr. Christoph Mayer (OFFIS Institute for
Information Technology). Christoph Mayer was the division manager of the R&D
division Business Information Management at the OFFIS Institute, and my rst
iiiiv Acknowledgement
superior. I would like to thank him for his support, his advice and the chance to
learn a lot about projects and leadership in science.
A very special thanks goes to the entire Information Systems and Simulation
group at the Goethe University Frankfurt am Main. The people that accompa-
nied me during my time at this group (in addition to Prof. Dr. Ingo J. Timm)
are Marion Terrell, Dr. Andreas Lattner, Tjorben Bogon, Yann Lorion, Pascal
Katzenbach and J org Dallmeyer. I have to thank all of them, and in particular
Dr. Andreas Lattner, for their support and advices.
Also, a very special thank you goes to my dear friend Mathias Uslar (OFFIS
Institute for Information Technology), a former colleague of mine. Especially in
my rst years of my endeavors in research he supported me with a lot of valuable
advice while in the last years he o ered much needed moral support.
I also have to thank Jens Oehlerking (University Oldenburg), who also sup-
ported me during the writing of this thesis.
Another person I met during my time as PhD student, who became a friend
that also helped me is Dr. Leif Meier (Proctor & Gamble Service GmbH). He got
me involved in the container terminal management problem and we have worked
collaboratively on this issue. Also, I would like to thank him for the planning
systems he provided to me.
I was also supported by Dr. David Sabel (Goethe Universit at Frankfurt) and
want to thank him for supporting the Latex template this thesis uses.
I would like to thank Dr. Michael Schwind (Goethe Universit at Frankfurt) for
the insightful discussions concerning combinatorial auctions.
I am very grateful to my former diploma students, who have contributed to as-
pects of my research that can partially be found in this thesis. In particular, I have
to mention Thomas Timmermann (Ernst & Young GmbH) and Zijad Kurtanovic
(Healy Hudson GmbH).
Also, I have to thank the students who participated in the practical course
\Praktikum Wirtschaftsinformatik und Simulation" in the winter term 2009/2010
for their e orts in developing planning systems and encapsulate them using web
services.
While I was employed at the Goethe University I was additionally supported
by GRADE, the Goethe Graduate Academy (former Otto Stern School) support-
ing me with subsidies for childcare. I received travel funding from the Herrmann
Willkomm-Stiftung in 2008 and 2009. Furthermore, I was supported by the EU-
COST Action Agreement Technologies. I received a grant for a short-term sci-Acknowledgement v
enti c mission (STSM) for a 10 day visit of the group of Prof. Dr. Sascha Os-
sowski in 2009. Finally, I was supported by the German Academic Exchange
Service (DAAD) with a short-term grant for doctorate students (grant number
D/09/48031). This scholarship allowed me to stay in Spain for three month in
2010.
For their ongoing support in so many di erent ways I would like to thank my
parents. I am very grateful for their support.
Especially in the last months of this thesis, my family supported me very much.
They had to put up with my stress and the fact that I was not there for three
month. Therefore, I am very grateful.
I would like to thank my two sons Philipp Ephraim and Simon Casimir for their
ongoing lecture they give to me to focus on the important aspects of life. They
were born during the time this thesis has been written, and know their father only
as a PhD. student, so far.
All these persons have in uenced me, and consequently, the way this thesis looks
like. But there would be no thesis in any way without the support and love of my
wife. I am deeply indebted to Jennifer for the advice and support she provided to
me during all this time. She brought me back on track when I lost sight of reality.vi AcknowledgementZusammenfassung
Reale Planungsprobleme, wie etwa die Produktionsplanung in einer Supply Chain,
sind komplex Planungsprobleme. Eine ublic he Strategie derart komplexen Proble-
men zuosen,l ist es diese Probleme in einfachere Teilprobleme zu zerlegen und diese
dann separat, meist sequentiell, zu osenl (divide-and-conquer Strategie). Dieser
Ansatz erlaubt die Erstellung von (suboptimalen) Pl anen fur eine Reihe von realen
Anwendungen und ist heute in den Organisationsstrukturen von gr o eren Un-
ternehmen institutionalisiert worden. Allerdings werden Abh angigkeiten zwischen
den Teilproblemen nicht ausreichend beruc ksichtigt, da die Partialprobleme se-
quentiell ohne Feedback gel ost werden. Die erstellten Teill osungen mussen deswe-
gen oft nachtr aglich koordiniert werden. Die Beispiele, die in dieser Arbeit genutzt
werden, sind die Koordination von Planungssysteme in der Produktion und Dis-
tribution von Gutern, die Koordination der Planungssysteme von Partner eines
Supply Chain, und die Koordination der Planungssysteme die im Rahmen des
Container Terminal Managements.
In dieser Arbeit wird die operative Koordination von existierenden Planungssys-
temen betrachtet. Dabei wird von einer inh arent gegebenen Problemstellung aus-
gegangen in der fur verteilte, abh angige Planungsentscheidungen automatische
Planungssysteme eingesetzt werden. Die Pl ane dieser Planungssysteme mussen
koordiniert werden, um die Realisierbarkeit dieser Pl ane unter Beruc ksichtigung
evtl. wechselseitiger Abh angigkeiten der Planungsprobleme zu gew ahrleisten. Je
nach gegebener Problemstellung kann dabei der Fokus neben der reinen gemein-
samen/parallelen Realisierbarkeit einer Menge von verteilt erstellten Pl anen auch
die Verbesserung einer Gesamtsystemperformance im Fokus liegen. Eine Koordi-
nation von Planungsprozessen auf der taktischen oder strategischen Ebene wird
in dieser Arbeit nicht betrachtet. Ein Beispiel fur derartige taktische oder stra-
tegische Koordinationsprobleme sind etwa die Auswahl eines Netzwerkes oder die
Zerlegung des Gesamtproblems in unterschiedliche Teilprobleme.
Im Rahmen der hier vorliegenden Arbeit wird zuerst ein Einblick in die Grund-
lagen von Planungssystemen und Systemen der verteilten