This project was done in collaboration with OCTOMATION robotic workforce formerly known as InMóTx and was a continuation of an earlier project made on 9th semester. The overall purpose was to increase the performance of an automated packaging system. Today OCTOMATION find it difficult to test modifications on the packaging system and therefore they want to do it virtually. The benefit is less resources spent on testing and the possibility to reproduce these tests.

It was decided to programme a simulation tool in C#. Software developing was studied and gave some methods on how to develope software and how to define the specifications of the software. Furthermore it gave some design guidelines for developing an user interface and how to document the architecture.

Analysing was commenced in order to study how to increase the performance of an automated packaging system. The speed by which the parts and boxes arrives in the packaging system is very high. It was concluded that scheduling methods which calculate the optimal solution can not be calculated fast enough. Therefore some heuristic prioritisation rules have been studied. These prioritisation rules will not give the optimal solution, but they will give a satisfactory solution. Furthermore some others rules and constrains are examined. Together with the prioritisation rules it is possible to create packaging strategies.

The simulation tool has been produced. The design guidelines from the analysis has been used which make the user interface user-friendly. It is possible to setup packaging strategies from the three basic rules of heuristic prioritsation rules, vertical prioritisation and minimum number of targets. These packaging strategies can then be tested in various packaging systems and under different workload circumstances. When the simulation is running, it is possible to see key performance indicators and graphs. When the simulation is completed, it is possible to process and analyse the results in MATLAB. The simulation tool has been validated in two steps. First the individual components have been validated to see if they operate as intended. Subsequently the entire simulation system were tested in order to validate that no errors would occur, when the components were put together. The simulation system performed with no errors and the validation was satisfactory.

The simulation tool was used to conduct a study on how various packaging strategies would perform under different circumstances. The results gave no precise answer to which strategy were best. However the combination called P1 were best at minimising the number of not-packed boxes. Furthermore it can be recommended to use FIFO in the first robot cell for boxes.
The results from the test were used to create a packaging strategy called P6, which uses all the best elements from the testings. It is recommended that further testing are conducted with this strategy.

This project has all in all resulted in a user-friendly and flexible simulation tool which OCTOMATION can use to develop and test new packaging strategies without using an excessive amount of resources.