Purpose:The primary purpose of the current thesis is to develop a virtual model using discrete event simulation (DES), which aims at supporting the decision-making process regarding automation solution proposals for SMEs. 

Method:The research approach is positivism, and it considers quantitative and empirical information. A literature search is conducted to generate a base for obtaining the theory required for the current report to answer the research questions. This search included the trace of relevant and reviewed topics regarding automation, discrete event simulation, and production lines. Then, a scenario simulation is designed and studied based on empiric knowledge and how automation would affect it, followed by a collection of information from the simulation iterations.

Findings& Analysis: Two scenarios are presented. One with a fully manually operated assembly line consisting of only human operators and a second scenario, a semi-automated assembly line that includes some robots in specific areas doing specific operations. The two scenarios are simulated to check to what extend the KPI’s and parameters improved between each scenario. The experiment result concludes that by introducing automation elements in the production line, there is an increase in the overall efficiency, throughput rate, and a considerable gap against humans in WIP.

Conclusions and recommendations: The results obtained from the experimentation in discrete event simulation software and study from previous research show that discrete event simulation has a significant contribution when considering a decision-making tool's role. Since it allows to understand and study the specific scenario by imitation and try different solutions in the same production system, it also allows studying several indicators from the scenarios to be checked to what extent they could be improved.

Delimitations: The current thesis includes several delimitations. First, it focuses only on an operational level. Also, this study consists of a specific type of product with many variants, and finally, there are only two scenarios studied: a fully manual scenario and a semi-automated scenario with the presence of robots.