Title Razvoj simulacijskoga modela za određivanje ključnih značajki linija obrade brodogradilišta Title (english) Development of a simulation model to determine the key performance indicators of shipyard’s processing lines Author Viktor Ložar Mentor Neven Hadžić (mentor)Committee member Jerolim Andrić (predsjednik povjerenstva)Committee member Tihomir Opetuk (član povjerenstva)Committee member Marko Hadjina (član povjerenstva)Granter University of Zagreb Defense date and country 2023-07-06, Croatia Scientific / art field, TECHNICAL SCIENCES Universal decimal classificationUDC ) 004 - Computer science and technology. Computing. Data processing 629.5 - Watercraft engineering Abstract Povijesni razvoj proizvodne industrije i njena uloga u suvremenom gospodarstvu
prikazani su u uvodnom odlomku, a u nastavku se ističe važnost brodograđevne industrije u
svijetu i u Hrvatskoj. Podpoglavlje Projektiranje i analiza proizvodnih sustava podcrtava smisao
proizvodnje te načine na koje se ona može unaprijediti, odnosno kategorizirati. Pregled
literature daje uvid u dosadašnje istraživanje proizvodne industrije uz istaknute nedostatke
vezane uz brodograđevnu industriju. Cilj ovog istraživanja je razvoj simulacijskog modela
kojim će se omogućiti evaluacija ključnih značajki serijskih Bernoullijevih proizvodnih linija.
Drugo poglavlje opisuje osnovne teorije skupova i Markovljevih lanaca na kojima se
temelji razvoj modela. Serijske proizvodne linije i linije s grananjem tokova materijala s
nepouzdanim strojevima i konačnim kapacitetom međuskladišta pri konstantom, homogenom,
vremenu izrade su opisane u trećem i četvrtom poglavlju. Pritom su detaljno objašnjeni prostor
stanja, ključne značajke, analitičko rešenje i nova metoda konačnih stanja.
Peto poglavlje definira uska grla za razne ključne značajke te opisuje projektiranje
proizvodnih sustava pomoću metode diferencijalne evolucije. U kratkom šestom poglavlju
opisuje se novonastali simulacijski model razvijen u programskom jeziku FORTRAN, a u
sedmome je prikazana njegova primjena u slučaju brodograđevnom proizvodnom sustavu.
Usporedba ključnih značajki dobivenih metodom konačnih stanja s analitičkim
pristupom potvrdila je točnost metode, a primjena je potvrdila mogućnost upotrebe modela u
stvarnom okruženju. Razvijena metoda konačnih stanja proširena s diferencijalnom evolucijom
pokazala se kao dobar alat za projektiranje novih proizvodnih postrojenja.
Abstract (english) The manufacturing industry is part of the global economy, and its development is closely
related to modern civilization's historical progress based on science and technology. The
introduction part of chapter one is highlighting these achievements and shows the relationship
between the manufacturing industry and the economic growth of a country. The first subheading
of chapter one shortly describes the historical tradition of the global shipbuilding industry and
shows which countries are the major players. The next subheading gives a short review of the
Croatian shipbuilding industry followed by a summary on design and analysis of production
systems. The main goal of this doctoral thesis as well as the hypothesis are outlined after the
literature review. The main goal is the development of a new simulation model which will
enable the evaluation of key performance indicators of serial Bernoulli lines. The hypotheses
address the possibility that a new model for serial and splitting production lines can be
established using a new method named the finite state method employed to calculate the key
performance indicators. As compared to conventional methods, such a new approach will result
in better design solutions of manufacturing systems, especially of the ship prefabrication and
fabrication lines.
The second chapter contains an overview the theory of sets and stochastic processes,
namely the Markovian chains. Therefore, the basic terms on Ven diagrams, set operators, set
algebra, probability theory and stochastic processes are outlined. The discrete time Markovian
chains are elaborated and explained more deeply, the Chapman-Kolmogorov equation is shown
and the eigenvalue problem is briefly presented.
Chapter three deals with the definition of serials lines and the necessary assumptions
needed to describe a Bernoulli distribution. In the first subheading the system state space is
presented and explained. The next subheading briefly presents the recently developed general
analytical solution to formulate the transition matrix. However, this approach is quite
challenging and time consuming due to the exceptional complexity of the system state space.
Therefore, the finite state method was developed based on the proportionality of the system
state space to reduce the evaluation time for lager system states. The aggregation method and a
simulation approach are briefly presented for the purpose of comparison with the finite state
method and the analytical approach. Such a comparison was done for the first time. The results
are highlighted in a short discussion with the conclusion that the new finite state method is
worth to be further developed for the case of splitting lines. The chapter four presents the definition of splitting lines and defines the analytical
approach. The finite state method for splitting lines was elaborated in the next subheading. The
finite state method for splitting lines was validated against the analytical approach and via
application case using various expressions for the key performance indicators. At the end, the
results prove that the finite state method is capable to model a splitting line and to calculate the
required key performance indicators.
The finite state method is employed to calculate the eigenvector for the entire system
state. This feature enables evaluation of bottlenecks for all key performance indicators. The
expressions for such operations are listen in the chapter five. At the end of this chapter the
differential evolution theory, is combined with the finite state method and a new design
methodology is presented.
The chapter six contains the scheme of the new simulation tool ShipProLab developed
in FORTRAN using the analytical approach and the finite state method.
In chapter seven an illustrative example of a prefabrication and fabrication production
line is presented. In this example, the developed theory is employed to determine the key
performance indicators and to calculate the time required to compete production of two typical
ship sections. Therefore, a brief description of the facility is given including the substitute
models. A new facility is designed using the differential evolution theory.
The last chapter summarizes the main conclusions of the research. The newly developed
finite state method is highlighted as a powerful tool to calculate the key performance indicators
for serial and splitting lines that, in combination with the differential evolution theory, enables
a sophisticated approach to design issues related to production system engineering.
Keywords
proizvodna industrija
brodograđevna industrija
Markovljevi lanac
serijske Bernoullijeve proizvodne linije
Bernoullijeve linije s grananjem tokova materijala
metoda konačnih stanja
ključne značajke
uska grla
metoda diferencijalne evolucije
Keywords (english)
manufacturing industry
shipbuilding industry
Markovian chains
serial Bernoulli lines
Bernoulli splitting lines
finite state method
key performance indicators
bottlenecks
differential evolution theory
Language croatian URN:NBN urn:nbn:hr:235:190236 Study programme Title: Mechanical Engineering and Naval Architecture Type of resource Text File origin Born digital Access conditions Open access Terms of use Created on 2023-07-12 11:45:16
