Integrating BIM models and AI models for cost estimation

master thesis no.: 278/II. GR-BIM

Faith Sharon Tangara (Author), Tomo Cerovšek (Mentor)

Abstract

Ocena stroškov ostaja ključni del projektnega vodenja, ki usmerja načrtovanje in sprejemanje odločitev skozi vse faze projektov. Tradicionalne metode so omejene zaradi kompleksnosti gradbenih elementov, zahtevane natančnosti in kratkih rokov. Kljub napredku informacijskega modeliranja zgradb (BIM),se pojavlja vse večja potreba po natančnejših, učinkovitejših in avtomatiziranih metodah izdelave ocen stroškov. Zato se magistrska naloga osredotoča na boljše razumevanje izdelave popisov in stroškovnikov na osnovi BIM ter potencial avtomatizacije. Cilj študije je prikazati potencial umetne inteligence za izboljšanje natančnosti, učinkovitosti in racionalizacijo procesov pri izdelavi popisov in oceni stroškov, kar bi lahko služilo kot ogrodje sistema za podporo odločanju popisovalcem in drugim deležnikom projektov. Pri raziskovanju potenciala umetne inteligence se osredotočamo zlasti na napredno analitiko in velike jezikovne modele LLM (angl. Large Language Medels), kot je ChatGPT, in na druge tehnologije v povezavi z BIM. Z integracijo LLM, BIM in AI bi lahko izdelavo popisov in oceno stroškov bistveno izboljšali tako v smislu kakovosti in učinkovitosti, medtem bi znatno zmanjšali obseg »ročnega dela«. Da bi lahko bolje izkoristili potencial umetne inteligence in BIM, bi bilo treba izboljšati interakcijo med LLM in modeli BIM ter uvesti napredno analitiko zgodovinskih podatkov kot učnih nizov, s katerimi bi bilo mogoče ogrodje aplicirati na razne vrste in velikosti projektov.

Keywords

civil engineering;master thesis;BIM;cost estimation;quantity take off;artificial intelligence;large language models;

Data

Language:	English
Year of publishing:	2023
Typology:	2.09 - Master's Thesis
Organization:	UL FGG - Faculty of Civil and Geodetic Engineering
Publisher:	[F. S. Tangara]
UDC:	004.946:69-047.74(043.3)
COBISS:	187626755
Views:	243
Downloads:	40
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	Slovenian
Secondary title:	Integracija modelov BIM in modelov AI za oceno stroškov
Secondary abstract:	Cost estimation remains a critical part of project management, guiding the budgeting and decision making throughout project phases. Traditional methods are limited due to complexity of building elements, required accuracy, and tight schedules. Despite recent advancements in Building Information Modelling (BIM) there is a growing demand for more accurate, efficient, and automated methods of cost estimation. Therefore, the thesis focus is given to understanding BIM-based Quantity Take-Off (QTO) and automation potential. The goal of the research is to demonstrate the potential of AI to enhance accuracy, efficiency, and streamline cost estimation processes that would support especially cost estimators, and also other project stakeholders. This research explores the potential of AI, specifically Large Language Models (LLMs) like ChatGPT, and other technologies in conjunction with BIM models. By integrating LLM, BIM, and AI cost estimation could be improved in terms of quality and efficiency, while manual work could be reduced significantly. To fully harness the potential of AI and BIM the interaction between LLMs and BIM models should be enhanced, as well as advanced analytics of historical data that would serve as learning sets should be put in place to allow for framework's applicability on diverse project types and scales.
Secondary keywords:	gradbeništvo;magistrska dela;GR;BIM;ocena stroškov;popisi;umetna inteligenca;veliki jezikovni modeli;
Type (COBISS):	Master's thesis/paper
Thesis comment:	Univ. v Ljubljani, Fak. za gradbeništvo in geodezijo
Pages:	XII, 75 str.
ID:	23066000