Utveckling av ramverk för driftstörningsreduktion i produktionsflöden : En fallstudie utförd på Husqvarna Group

Sammanfattning

To achieve an efficient flow in a production system, the degree of resource utilization is one ofthe essential components. A common way of measuring resource utilization is through the keyperformance indicator Overall Equipment Effectiveness (OEE). One of the categories used incalculating OEE is availability, where unplanned downtime negatively affects the OEE-value.To reduce unplanned downtime, the reliability of the production flow is one of the key aspectsto consider. There is a lack of frameworks that clearly and practically demonstrate how systematicimprovement work should be carried out to reduce unplanned downtime in a real productionenvironment. While partial methods and generalized frameworks exist, they do not show howa concrete systematic, and at the same time practical method can be structured. In this study, the Production Disturbance Analysis Framework (PDAF) was developed withthe purpose of increasing reliability in the part of a production system responsible for thelargest amount of unplanned downtime. The framework is intended to be concrete,comprehensible, practically applicable and flexible. PDAF is based on widely accepted and well-established industrial theories and methods, such as the Ishikawa Diagram and Fault TreeAnalysis. To evaluate the functionality of PDAF in a production environment, it was implemented in acase study at Husqvarna Group´s automated assembly cell for consumer chainsaws. Theframework proved to be practically applicable, although challenges related to the complexity ofthe assembly cell were identified. A current state analysis of the production cell was conducted,during which data was collected and subsequently used within PDAF throughout the analysisprocess. The work resulted in a proposed solution that was implemented in active assemblyoperations on the production line to identify the effect of the improvement. Finally, the reportdiscusses validity and reliability, as well as the conclusions drawn in response to the researchquestions.