Projects / Programmes
Analysis of failures that were detected at technical inspection procedures by using conventional statistical methods and data mining methods
| Code |
Science |
Field |
Subfield |
| 2.11.04 |
Engineering sciences and technologies |
Mechanical design |
Machine parts, engines and devices |
| Code |
Science |
Field |
| T003 |
Technological sciences |
Transport technology |
| Code |
Science |
Field |
| 2.03 |
Engineering and Technology |
Mechanical engineering |
technical inspection of vehicle, technical failures, failure analysis per region, failure analysis per contractor, multivariate analysis, data mining
Researchers (7)
| no. |
Code |
Name and surname |
Research area |
Role |
Period |
No. of publicationsNo. of publications |
| 1. |
16334 |
PhD Jernej Klemenc |
Mechanical design |
Head |
2019 - 2022 |
782 |
| 2. |
50821 |
PhD Tadej Kocjan |
Mechanical design |
Researcher |
2020 - 2022 |
45 |
| 3. |
13469 |
PhD Marko Nagode |
Mechanical design |
Researcher |
2019 - 2022 |
812 |
| 4. |
26561 |
PhD Simon Oman |
Mechanical design |
Researcher |
2019 - 2022 |
254 |
| 5. |
29047 |
PhD Domen Šeruga |
Mechanical design |
Researcher |
2019 - 2022 |
337 |
| 6. |
32031 |
PhD Urša Šolinc |
Mechanical design |
Researcher |
2019 - 2021 |
28 |
| 7. |
39298 |
PhD Peter Zobec |
Mechanical design |
Researcher |
2019 - 2022 |
36 |
Organisations (1)
Abstract
In the project an analysis of failures and faults will be carried out on the basis of data from technical inspections of road vehicles.
The objectives of the project are based on the requirements from the call of proposals:
1.) Collecting data on the number of performed technical inspections of road vehicles at individual contractors.
2.) Estimation of the vechicles with no failures or faults for each contractor.
3.) Determination of the three most frequent failures or faults detected at technical inspections.
4.) Analysing failure data for different regions of Slovenia.
5.) Analysing failure data for the contractors.
In the project primary and secondary objectives are defined. Primari objectives are: i.) for a period of two consecutive years the number of technical inspections for each road-vehicle category is determined; ii.) analysis of vehicle's techical adequacy and detected failures is carried out; iii.) a comparison with a developed western country (i.e. Germany, because a lot of the relevant data is publically available) is made. By meeting the primary objectives an ensemble of recommendations can be set up in order to improve technical adequacy of slovenian vehicles.
The secondary objective is a comparative analysis of failure distributions for the Slovenian regions and contrators, which are allowed to perform technical inspections of road vehicles. If significant discrepancies are found, recommendations for their reductions will be defined.
To meet the project objectives the project activities are divided into six phases that are performed in a logical sequence with a certain overlapping:
1.) Initial acquisition of data on road-vehicle technical inspections from the contractors.
2.) Analysis of the initial data set and search for the missing and incomplete data.
3.) Definition of recommendations and methods for improving input-data quality.
4.) Acquisition of data for the final statistical analysis and data mining.
5.) Intermediate and final data analysis.
6.) Dissemination of project results.
In the initial project phases the quality of data on mechanical failures and faults , which are gathered from the contractors, will be estimated. Due to a big amount of data that are acquired from a number of different sources some issues related to the missing or incomplete data are expected. If the data quality varies among the different contractors the outcomes of the statistical analysis may not be reliable. For this reason it is important that a procedures and methodologies are developed that will enable adequate treatment of samples with missing or incomplete data.
The acquired data on road-vehicle mechanical failures and faults Will be then analysed with different conventional statistical and unconventiona data mining methods. Further on statistical discrepancies of detected failures and faults will be analysed on the regional- and contractor-wise basis. In this manner local peculiarities related to the vehicle age spectrum will be identified. Contractor-based data will be compared for different regions and Slovenia as a whole country to identify significant differences, if they exist. If significant differencis do exist we will try to find the root casue with aditional analyses.
The final results from data analyses on the road-vehicle failure data will be compared with publically available data from the developed western country (Germany). Similarity of the technical-failure distribution and vehicle's age-related failures will be estimated.
The following results, which are related to the primary objectives, will be available at the end of the project:
- The state of the technical inspections in Republic of Slovenia will be identified for two consecutive years.
- The state of road-vehicle technical adequacy will be identified for the complete vehicle-age spectrum. Typical failures and faults will be correlated to the vehicle's age.
- A comparison of vehicle state with Germany will be available from tw
Significance for science
The members of the project group are involved with an application and development of advanced statistical analyses and data mining for twenty years. Until this project they applied these methods mostly for solving tecnical problems with limited data sets or (to a smaller extent) for analysing the real-estate market in Slovenia. In the latter reseearch the missing or incomplete data were the bigeest problem that was solved. In this project we exdpect that the statistical and data mining methods Will be applied on the huge data sets (big-data problem), since it is evident form the statistics of SURS that approximatelly one and half milion of vehicles are registered in Slovenia. This meand that there is a few hundreds of thousands of technical inspection processes being carried out each year when considering the inspection frequencies as defined in ZMV-1 law from the year 2017. Due to a rather large number of companies that are allowed to perform technical inspections of vehicles it is expected that a large number of missing and incomplete data will be present in the gathered data. It will be a special challenge of this project how to model the missing data. The appropriate solution of this challenge can represent a clear scientific added value of the project. The results of the statistical analyses and data mining can further represent a scientific added value, because statistical distribution of mechanical failures among the regions and failure patterns are not known at the time of project application. Besides it is very difficult (with some exceptions, i.e. Germany) to find reliable data on typical mechanical failures and their frequencies that are discovered during technical inspections. For this reason we expect that the project results will not only contribute to the improved traffic safety in Slovenia, but will also result in scientific publications that will be published in international journals and presented at international conferences.
Significance for the country
A direct added value pf the project will be for all the companies that are involved in the vehicle's technical inspection business. The benefit for these companis will be that a technical supervision processes among different players could be made more uniform (despite the fact that they are presribed by the law). Namely, if significantly different failure distributions will be discovered for different companies, this may imply that the presribed processes are not being executes equally and there is a room for some optimisation.
The broader economic meaning of the project is that it would be possible for the owners of commercial vehicles (combined vehicles, trucks, buses) to gain an information on the increasing probability of failures for the aeging vehicles. With this information they will be able to estimate the profitable commercial-usage period of the vehicle more accuratelly, which can lead to a better planning of fleet replacement times for the vehicles.
Most important scientific results
Most important socioeconomically and culturally relevant results
Annual report
2020
