We describe the development of an agent-based Decision Support System in the spunlace nonwovens production industry. This work constitutes the first pilot system that models and evaluates major products costs by using Activity Based Costing methodology in INOTIS enterprise. One of the main aspects addressed in the decision-making process is the knowledge management of the most frequently failures of machines as the card, aqua jet etc. This paper mainly focuses on proposing an agent model composed of six agents namely Resource Agent (RA), Cost Agent (CA), COordinator Agent (COA), Data Manager Agent (DMA), Maintenance Agent (MA) and Scheduling Decision selection Agent (SDA). Not only was possible to support the necessary decisions, but several models were also displayed so that production manager or decision -maker could support new problems.