Analiza CAD aktivnosti u višekorisničkom CAD okruženju

master's thesis

• Title (english): Analysis of CAD activities in a multi-user CAD environment
• Author: Lovro Sever
• Mentor: Stanko Škec (mentor)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2024-07-16, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Mechanical Engineering

Abstract

Upotreba CAD (eng. Computer Aided Design) alata važna je u kasnijim fazama razvoja proizvoda. Korištenjem CAD alata, razvojni tim kreira digitalni model proizvoda. Tradicionalni CAD alati omogućuju asinkroni rad na digitalnim modelima čime je otežana kolaboracija između članova tima. Razvojem računarstva u oblaku omogućeni je razvoj višekorisničkih CAD alata. Jedan od vodećih višekorisničkih CAD alata smještenih u oblaku (eng. Cloud-based software) je Onshape® tvrtke PTC koja omogućuje sinkroni rad više članova tima na istom modelu. Onshape® omogućuje kontinuirano praćenje aktivnosti članova razvojnog tima prilikom CAD modeliranja. Za potrebe rada korišteni su podaci o CAD aktivnostima studentske udruge koja se bavi izradom raketa. U praćenom razdoblju postoje zapisi o 18453947 aktivnosti proizašlih od strane 491 korisnika (članova tima). Na samom početku rada provedeno je istraživanje postojeće literature vezanu za analizu CAD aktivnosti što je uvelike pomoglo definiranju pristupa samoj analizi podataka. Analiza aktivnosti podijeljena je u tri dijela. U sklopu prvog dijela generirane su grupe korisnika i Onshape dokumenata s obzirom na broj zabilježenih aktivnosti, te je provedena analiza aktivnosti za svaku generiranu grupu. U drugom dijelu analize, za potrebe klasteriranja dokumenata i korisnika, korišteni su K-Means i GMM algoritam. Pomoću oba algoritma provedeno je klasteriranje podataka o aktivnostima te su navedeni rezultati analize. Struktura CAD aktivnosti grupa dokumenata i korisnika ovisi o pristupu grupiranja istih. Generirani klasteri pomoću dva različita algoritma daju drugačije rezultate ističući pojedine klastere (s obzirom na strukturu CAD aktivnosti) od ostalih klastera U sklopu trećeg dijela analize odrađeni je pregled kolaboracije unutar tima. Korisnici (Članovi tima) Analiziranjem dinamike promjena mreže kolaboracije utvrđeno je postojanje podtimova koji rade zajedno, ali ti podtimovi variraju na mjesečnoj bazi.

Abstract (english)

The usage of CAD (Computer-Aided Design) tools is important in the later stages of product development. With CAD tools, the design team can create digital models of the product. Traditional CAD (single-user CAD) tools enable work from one PC and one user at a time on a single CAD model. This restriction makes collaboration more difficult.
The development of cloud computing has enabled the creation of multi-user CAD tools. One of the leading multi-user CAD tools is Onshape, a PTC company product that allows more than one user to work on a CAD model simultaneously. Onshape also enables tracking of user activity using the Audit Trail module.
For this master’s thesis, a dataset of CAD activity from a student association involved in rocket design is used. During the tracking period, 18,453,947 CAD activity were recorded from 491 users. At the beginning of this work, a survey of the existing literature related to the analysis of CAD activities was conducted, which greatly helped define the approach to data analysis.
The analysis of CAD activity is divided into three parts. The first part involves grouping users and Onshape documents based on the number of recorded timestamps. Each created group is analyzed separately. In the second part, GMM (Gaussian Mixture Model) and K-Means clustering algorithms are used to generate clusters of documents and users. Clustering of data for CAD activity is performed using each algorithm, and the results are compared. Each approach to data grouping provides different insights into the structure of CAD activity. The use of different clustering algorithms results in the creation of some clusters that differ in their approach to CAD activity.
The last part is the analysis of user collaboration. After analyzing the dynamics of the collaboration network, the existence of sub-teams was established, but only on a monthly basis.

• Keywords: analiza CAD aktivnosti; identifikacija obrazaca; klasteriranje; Onshape Audit trial
• Keywords (english): CAD activity analysis; pattern identification; clustering; Onshape Audit trial
• Language: croatian
• URN:NBN: urn:nbn:hr:235:481129
• Study programme: Title: Mechanical Engineering; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka strojarstva (magistar/magistra inženjer/inženjerka strojarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Embargoed access; Embargo expiration date: 2026-07-12
• Created on: 2024-07-15 01:00:02