Past, present and future, to realize the aim of product CTQS (i.e., lower cost, faster time to market, higher quality and better service) with manufacturing intelligence, few manufacturers have no longer engaged in product related production decision support problem (P-DSP). However, P-DSP solving (P-DSPS) is a multi-criteria decision-making problem, which is context sensitive in solution objects-attributes and chaos in the decision process of manufacturing knowledge collaboration and reuse. To alleviate these limitations, this paper presents a novel triple deep workflow model for P-DSPS. Driven by a wicked task query, the proposed workflow of P-DSPS (WP-DSPS) has the function to retrieve similarity-based alternatives from domain knowledge driven solution flow (KSF) and to evaluate with expert knowledge collaboration from knowledge driven decision flow (KDF) based on utility theory under the task event driven control flow (ECF) strategy and operation logic. In the view of alternative adaption, a domain knowledge ontology-based degree of similarity (DoS) determines the P-DSPS alternatives width, a utility function-based degree of decision (DoD) determines alternatives quality, and a belief-based knowledge fusion technique is used to synthesize decision conflicts with a consensus degree (CD). To support the proposed models, a workflow-based system prototype is proposed and validated in two case studies.