國立虎尾科技大學 |

Cost-Safety-Aware Inspection Strategy for Truck Fleets with Limited Historical Data.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Cost-Safety-Aware Inspection Strategy for Truck Fleets with Limited Historical Data./
Author:	Shi, Ying.
Description:	1 online resource (143 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
Contained By:	Dissertations Abstracts International85-11B.
Subject:	Computer science. -
Online resource:	click for full text (PQDT)
ISBN:	9798382372341

Cost-Safety-Aware Inspection Strategy for Truck Fleets with Limited Historical Data.
Shi, Ying.

Cost-Safety-Aware Inspection Strategy for Truck Fleets with Limited Historical Data. - 1 online resource (143 pages)

Source: Dissertations Abstracts International, Volume: 85-11, Section: B.

Thesis (Ph.D.)--Carnegie Mellon University, 2024.

Includes bibliographical references

Heavy-duty trucks are a significant segment of the population involved in fatal accidents. In order to prevent crashes caused by vehicle malfunctions, effective inspection planning is essential. This research aims to develop a reliable predictive inspection planning strategy that can predict the future condition of the vehicle and to develop an inspection plan that can identify risky components with a few inspection costs and time. However, the limited historical data decrease the reliability of predictive inspection planning. To address this issue, this research developed data augmentation methods to generate synthetic data to fill in the limited historical data. The research also explores the potential to integrate humans and machines for more reliable inspection planning by handling data limitations from another perspective. This research addresses the challenges that: (1) the data augmentation is required to generate synthetic data similar to data in the real world, (2) the need for a comprehensive summary of characteristics of humans and machines in inspection planning, especially their advantages and limitations; (3) the need for integrating human knowledge into machine learning models with utilizing their advantages and avoiding limitations. To address the challenges, the researcher: (1) proposed a data augmentation method adapted to brake inspection planning considering the brake deterioration mechanism; (2) compared human and machine performance in inspection planning and summarized the scenarios in which human or machine perform better; (3) obtained a comprehensive summary of the advantages and limitations of human and machine inspection planning; and (4) This research proposed a framework for a human-machine collaboration mode that combines human and machine advantages and avoids their limitations. The outcomes of the research are expected to provide commercial fleets with reliable interpretation-based inspection suggestions, which can ensure vehicle safety with minimum inspection resources.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798382372341Subjects--Topical Terms:

573171
Computer science.
Subjects--Index Terms:

Data imputationIndex Terms--Genre/Form:

554714
Electronic books.

Cost-Safety-Aware Inspection Strategy for Truck Fleets with Limited Historical Data.
LDR:03432ntm a22003977 4500 001 1151262
005 20241104055840.5
006 m o d
007 cr bn ---uuuuu
008 250605s2024 xx obm 000 0 eng d
020 $a 9798382372341
035 $a (MiAaPQ)AAI31244101
035 $a AAI31244101
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Shi, Ying. $3 1194902
245 1 0 $a Cost-Safety-Aware Inspection Strategy for Truck Fleets with Limited Historical Data.
264 0 $c 2024
300 $a 1 online resource (143 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
500 $a Advisor: Tang, Pingbo.
502 $a Thesis (Ph.D.)--Carnegie Mellon University, 2024.
504 $a Includes bibliographical references
520 $a Heavy-duty trucks are a significant segment of the population involved in fatal accidents. In order to prevent crashes caused by vehicle malfunctions, effective inspection planning is essential. This research aims to develop a reliable predictive inspection planning strategy that can predict the future condition of the vehicle and to develop an inspection plan that can identify risky components with a few inspection costs and time. However, the limited historical data decrease the reliability of predictive inspection planning. To address this issue, this research developed data augmentation methods to generate synthetic data to fill in the limited historical data. The research also explores the potential to integrate humans and machines for more reliable inspection planning by handling data limitations from another perspective. This research addresses the challenges that: (1) the data augmentation is required to generate synthetic data similar to data in the real world, (2) the need for a comprehensive summary of characteristics of humans and machines in inspection planning, especially their advantages and limitations; (3) the need for integrating human knowledge into machine learning models with utilizing their advantages and avoiding limitations. To address the challenges, the researcher: (1) proposed a data augmentation method adapted to brake inspection planning considering the brake deterioration mechanism; (2) compared human and machine performance in inspection planning and summarized the scenarios in which human or machine perform better; (3) obtained a comprehensive summary of the advantages and limitations of human and machine inspection planning; and (4) This research proposed a framework for a human-machine collaboration mode that combines human and machine advantages and avoids their limitations. The outcomes of the research are expected to provide commercial fleets with reliable interpretation-based inspection suggestions, which can ensure vehicle safety with minimum inspection resources.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Environmental engineering. $3 557376
653 $a Data imputation
653 $a Data sparsity
653 $a Human knowledge
653 $a Human-machine collaboration
653 $a Predictive inspection
655 7 $a Electronic books. $2 local $3 554714
690 $a 0543
690 $a 0984
690 $a 0775
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Carnegie Mellon University. $b Civil and Environmental Engineering. $3 1190439
773 0 $t Dissertations Abstracts International $g 85-11B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31244101 $z click for full text (PQDT)