Extraction method of implicit knowledge from ship design drawings based on Apriori algorithm

【Objective】 Addressing the challenge of implicit knowledge being difficult to express in ship design drawings, this paper proposes a data mining method based on the Apriori association rule algorithm, aiming to uncover potential design patterns from existing design data. 【Methods】 Using a set of 400 ship fire protection design drawings as the research subject, the Apriori algorithm was employed to extract key indicators such as fire zone area, safety exit width, and firewall fire resistance rating. A transactional dataset was constructed, and frequent item sets and strong association rules were mined. 【Results】 The results demonstrate that the proposed method can effectively identify implicit knowledge in design drawings. Frequent item sets and association rules were discovered, including: service desks and escape exits are closely correlated in layout, fire zone area accounts for 10%–30% of the total area, and firewall fire resistance rating increases with larger building area. 【Conclusion】 This study validates the effectiveness of data mining technology in extracting implicit knowledge in ship design. The discovered rules can provide data-driven support for optimizing fire protection design and promote the transition of the design process from experience-driven to data-driven.