Crowdsourced manufacturing through a platform-driven manner has been observed as an emerging trend towards Industry 4.0 by paving the way of delivering Manufacturing-as-a-Service (MaaS). It utilizes a cyber platform and crowdsourcing to reach external partner’s manufacturing knowledge and resources while allowing companies to focus on their core competencies. It addresses an underlying logic that maximizing the reuse of resources by searching similarities among prolific product, process, and manufacturing resources varieties. It also challenges traditional logistic service for manufacturing industries by expanding a simple material flow to a complex networked and fluctuate one. Cross-docking has been widely recognized as a logistic solution to complex material flow by splitting service routes to pickups and deliveries for maximizing vehicle reuse. It adopts a platform-driven strategy by exchanging loads at the cross-docking. This study formulates the logistic service problem in platform-driven crowdsourced manufacturing as an Open Vehicle Routing Problem with Cross-Docking (O-VRPCD), which integrates logistic solution provider crowds into the manufacturing service process. This study considers the logistic provider as a capacitated homogeneous vehicle starting at various pickup points and times in a logistic service. The vehicles are scheduled in a route to visit service requesters synchronously and arriving cross-dock center simultaneously for load exchanging. Thus, this study formulates a mixed-integer programming (MIP) model for OVRPCD to minimize a total cost of crowdsourcing fleet, which considers logistic solution provider hiring costs and vehicle operation costs. A branch-and-price (B&P) algorithm is proposed to solve this problem using Pulse Algorithm-based column generation.