Odkrivanje kršitev znotraj sistema za registracijo delovnega časa in kontrolo pristopa Time&Space

VITEZ, DENIS (author), Sadikov, Aleksander (mentor)

Namen diplomskega dela je predstavitev rešitve odkrivanja kršitev v sistemu za registracijo delovnega časa T&S. Klasične rešitve odkrivanja kršitev slonijo na vnaprej definiranih pravilih, ki so pogosto del samega sistema in se težko spreminjajo. Pravila so po navadi odvisna tudi od lokalnih zakonov, ki veljajo v posameznih državah, saj težko najdemo državo, kjer bi se delovna zakonodajo popolnoma ujemala s katero izmed drugih držav. Namen tega diplomskega dela je zato predstaviti način, ki ne temelji na vnaprej definiranih pravilih, ampak na vedenjskih navadah uporabnikov in odstopanjih od običajnih vzorcev obnašanja. V sklopu diplomskega dela smo razvili spletno aplikacijo, ki na podlagi podatkov iz obstoječega T&S sistema operaterju predlaga možne kršitve.

• Language: Slovenian
• Keywords: registracija delovnega časa, Time&Space, osamelci
• Work type: Bachelor thesis/paper
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2016
• PID: 20.500.12556/RUL-85561
• Publication date in RUL: 16.09.2016

Secondary language

• Language: English
• Title: Detecting violations using time and attendance control system Time&Space

Abstract

The purpose of this thesis is to present solution for detection of violations in time and attendance system T&S. Classical solutions for detecting violations are based on predefined rules which are typically hard coded into systems and are difficult to change. Rules are usually dependant on local laws that are enforced in individual countries. It is almost impossible to find two countries that would have identical work related laws. Therefore this thesis presents a way that is not based on hard coded rules but it uses worker habits and deviations from their normal patterns. To present the findings of this thesis web application was developed that presents operator with possible violations.

• Keywords: time and attendance system, Time&Space, outlier detection