Title Sustav za podršku odlučivanju u planiranju obnove povijesnih cestovnih mostova : disertacija Title (english) A decision support system for the historic road bridges rehabilitation planning : doctoral thesis Author Katarina Rogulj Mentor Nikša Jajac (mentor)Committee member Deana Breški (predsjednik povjerenstva)Committee member Jure Radnić (član povjerenstva)Committee member Ivan Marović (član povjerenstva)Granter University of Split Defense date and country 2018-09-25, Croatia Scientific / art field, TECHNICAL SCIENCES Scientific / art field, TECHNICAL SCIENCES Universal decimal classificationUDC ) 624/625 - Civil and structural engineering. Civil engineering of land transport. Railway engineering. Highway engineering Abstract Disertacijom je iskazano istraživanje uspostave sustava za podršku odlučivanju (SPO) u planiranju obnove povijesnih cestovnih mostova. Najprije su dane teorijske postavke i dosadašnja istraživanja SPO-a, objašnjenje pojma povijesnih cestovnih mostova, njihovih karakteristika i važnosti te potencijalnih metoda obnove. Nadalje je oblikovan i opisan novoformirani SPO te njegovi osnovni elementi: baza podataka, baza znanja, baza modela i dijalog. Težište istraživanja je u oblikovanju baze modela, unutar koje su razvijena četiri nova modela s kojima se ostvaruje glavni cilj istraživanja. Predstavljena je struktura baze podataka te su prikupljeni adekvatni podaci i potrebna znanja za formiranje baze znanja. Tako prikupljeni podaci i znanja utjecali su i na proces formiranja modela u bazi modela. Kroz tri razine odlučivanja, uspostavljena je interakcija među navedenim osnovnim elementima SPO-a te je predstavljena njegova funkcionalnost u teorijskoj razradi. Na operativnoj razini odlučivanja prikupljeni su podaci za bazu podataka kao što su: nazivi i lokacije mostova, podaci o karakteristikama mostova, metodama obnove, potrebama u smislu prostorno-funkcionalnih i financijskih upravljačkih ograničenja te ekspertnom vizualnom načinu procjenjivanja stanja mostova. Na taktičkoj razini definiran je način formiranja ciljeva i relevantnih kriterija za planiranje metode obnove pojedinačnog mosta. Izrađen je novi model koji pomaže pri donošenju taktičkih odluka o izboru metode obnove, baziran na tim kriterijima i višekriterijalnom pristupu uz korištenje metoda AHP i PROMETHEE. Nakon oblikovanja modela za taktičku razinu koja se bavi planiranjem obnove pojedinačnog mosta prešlo se na oblikovanje modela za planiranje obnove skupa mostova. Tako je predstavljena metodologija uspostave kriterija vrednovanja potrebnih za realizaciju novo razvijenog modela rangiranja povijesnih cestovnih mostova prema prioritetu za aktivnost obnove. Ovo su strateški kriteriji te se kao rezultat primjene višekriterijalnih metoda (AHP, PROMETHEE i GAIA) dobio ukupni poredak analiziranih mostova. Među navedenim kriterijima posebno treba istaknuti doprinos ostvaren kroz definiranje kriterija kojim se opisuje stanje povijesnog mosta. Stoga je za tu potrebu izrađen novi pokazatelj nazvan indeks procjene stanja povijesnih cestovnih mostova (IPSPCM). Vrijednost, koja predstavlja ocjenu po tom kriteriju, iskazuje se indeksom koji je rezultat novo formiranog modela procjene stanja povijesnih cestovnih mostova. Ovaj model zasnovan je na kombiniranju definiranog načina provedbe ekspertne vizualne procjene stanja mosta, modificirane neizrazite (fuzzy) logike i odabrane višekriterijalne metode (AHP). Ukupni poredak mostova, dobiven kao rezultat modela rangiranja povijesnih cestovnih mostova prema prioritetu za aktivnost obnove, uz definiranje potrebnih prostorno-funkcionalnih i financijskih ograničenja te njihove kombinacije, korišten je za realizaciju novo oblikovanog modela izbora upraviteljskog pristupa. Kao rezultat analiziranog modela, primjenom višekriterijalne metode PROMETHEE V, dobio se set mostova za obnovu na godišnjoj razini. Glavni doprinos disertacije je razvoj SPO-a kojim se pomaže krajnjem donositelju odluke provoditi planiranje obnove povijesnih cestovnih mostova na sustavan, sistematičan i organiziran način. Validacija modela SPO-a napravljena je za slučaj povijesnih cestovnih mostova na području Splitsko-dalmatinske županije, izgrađenih za vrijeme ili prije Austro-Ugarske Monarhije.
Abstract (english) In this dissertation a study of a decision support system (DSS) establishment in the rehabilitation planning of historic road bridges is given. The theoretical background of DSS, historic road bridges explanation, their characteristics and importance and potential rehabilitation methods are presented at a beginning of dissertation. Furthermore, the newly formed DSS has been designed and described as well as its basic elements: database, knowledge base, model base and dialogue. The research focus was in designing the model base, within which four new models are developed. The structure of the database has been presented and adequate data are collected as well as required knowledge to form a knowledge base. The collected data and knowledge influenced the modeling process in the model base. Through three levels of decision-making, an interaction between the mentioned basic elements of the DSS was established and its functionality was introduced. At the operational level, data were collected for databases such as: name and location of bridges, bridge characteristics, rehabilitation methods, spatial-functional and financial constraints and expert visual bridge condition estimation. At the tactical level, the way of forming goal three and relevant criteria for an individual bridge rehabilitation method selection are defined. A new model has been developed to assist in making tactical decisions in the selection of rehabilitation methods, based on defined criteria and a multicriterial approach using the AHP and PROMETHEE methods. After design of the model on tactical level, the model for rehabilitation priority ranking of all historical bridges on strategic level is developed. Thus, the methodology for criteria establishing, necessary for the realization of the newly developed model, is presented. These are the strategic criteria and as a result of the multicriteria methods (AHP, PROMETHEE and GAIA) application, the overall ranking of the analyzed bridges was obtained. Among the above mentioned criteria, is the criterion that presents historic road bridge condition evalutaion. Therefore, a new indicator called the Index of Historical Road Bridge Condition Rating (IHRBCR) was developed. The value of IHRBCR is the result of a newly established model of historic road bridges condition estimating. This model is based on combining a defined way of implementing expert visual bridge condition rating, modified fuzzy logic and multicriteria method (AHP). The overall rank of bridges, obtained as a result of the model for rehabilitation priority ranking of historic road bridges, with the definition of the necessary spatial-functional and financial constraints and their combination, was used for the realization of the newly developed model for manager's approach selection. As a result of the analyzed model, using the multicriteria method PROMETHEE V, a set of bridges for rehabilitation was obtained on an annual basis. The main goal of the dissertation is to develop a DSS that will assist the final decision-maker to carry out the rehabilitation planning of historic road bridges in the systematic and organized manner. The model was validated on the case of historic road bridges in the area of Split-Dalmatia County, built during or before the Austro-Hungarian Monarchy.
Keywords
sustavi za podršku odlučivanju
višekriterijalna analiza
višekriterijalne metode
ekspertni sustavi
planiranje obnove
povijesni cestovni mostovi
IPSPCM
Keywords (english)
decision support systems
multicriteria analysis
multicriteria methods
expert systems
rehabilitation planning
historic road bridges
IHRBCR
Language croatian DOI https://doi.org/10.31534/DocT.046.RogK URN:NBN urn:nbn:hr:123:265043 Study programme Title: Civil Engineering Catalog URL https://library.foi.hr/lib/knjiga.php?B=422&H=&E=&V=&lok=&zbi=&item=14086 Type of resource Text Extent [10], 189 listova : tabele, ilustr. u bojama ; 30 cm File origin Born digital Access conditions Open access Terms of use Created on 2018-10-17 11:29:44
