Options
2013
Conference Paper
Titel
Automated test case generation from complex environment models for PLC control software testing and maintenance
Abstract
In the industrial automation domain, Programmable Logic Controllers (PLC) control production plants; and nowadays, PLCs mainly operate by means of embedded software. Of late this control software is increasing in size and especially in importance because they are employed in safety critical scenarios. Hence, a thorough testing of PLC control software is necessary. However, till today testing is one of the weakest points in the current development process. This is mainly because testing in the automation domain is a human intensive activity; and such manual testing is usually unproductive, often inadequate, and requires high efforts. A solution to the above problem is to automate the test case creation, execution and evaluation process. In this work, we present our methodology and demonstrate our fully functional framework developed exclusively for the purpose of generating tests for PLC testing. In the proposed framework, we test the control logic of the PLCs using the test cases generated from environment models. Testing PLCs using environment models provide the opportunity to exercise realistic plant scenarios on the control logic of the PLC, even before its actual commissioning at the real production plant. Test case generation from models is not new and several different approaches have been proposed overtime to generate test cases from models. However, the problem with existing research works is that most of the proposed approaches deal with test case generation from discrete, deterministic models and do not handle complex models. Unfortunately, the industrial automation environment model is neither discrete nor deterministic. Rather it shows complex characteristics. Thus, for our purpose of testing PLC software using tests from environment models, we have developed a novel approach to generate test cases from such complex environment models.