In an order-driven supply chain (e.g. perishable product supply chain) consisting of a single manufacturer and multiple customers (i.e. downstream customer enterprises), the manufacturer determines its production capacity (e.g. number of workers) at a certain time based on the orders that have been already received and the predicted orders to come, and fulfills orders beyond the capacity by increasing it with higher production cost rate. In this paper, we introduce an early order commitment (EOC) strategy into the order-driven supply chain, wherein each customer enterprise places orders either before (i.e. early orders) or after (i.e. regular orders) the determination of manufacturer's production capacity independently without knowing others' order timings. Towards this end, we develop an analytical model for a grafted seedling supply chain as a case study to quantify the manufacturer's production capacity and customer enterprise's optimal order quantity, respectively. To estimate the production cost rate, a simulator was developed mimicking nursery and grafting operations together with related material handling activities. The experiment results indicate that EOC is beneficial to the entire supply chain but increases customer enterprise's demand forecasting error, and each customer's purchase cost rate is affected by both its own and others' ordering times.