本文讨论了流程工业智能制造对过程系统工程(PSE)研究人员提出的挑战。现有的研究在实现全厂和全站点优化方面已经取得了很大进展,进行基准化测试能够增加说服力。本文进一步讨论了过程系统工程师在开发可用工具和技术时遇到的技术性挑战,包括灵活性和不确定性,响应性和敏捷性,鲁棒性和安全性,混合物性质和功能的预测,以及新的建模和数学范式。利用大数据进行智能化开发来驱动系统灵活性需要面对新的挑战,例如,如何在漫长又复杂的供应链中确保数据的一致性和机密性。建模方面也存在很多挑战,涉及如何对所有的关键技术进行恰当的建模,特别是健康、安全和环境方面,需要在特定地点对微小却关键的量进行准确预测。对环境方面的关注要求我们紧密跟踪所有的分子种类,以便于它们能被最佳地用于创造可持续的解决方案。而源自于新型个性化产品的破坏性商业模式对环境的影响则难以预测。 This article discusses the challenges that process industrial intelligent manufacturing poses to process system engineering (PSE) researchers. Existing research has made great strides in achieving plant-wide and site-wide optimization, and benchmarking can increase persuasion. This article further discusses the technical challenges encountered by process system engineers in developing available tools and techniques, including flexibility and uncertainty, responsiveness and agility, robustness and security, predictions of the nature and functionality of mixtures, and new Modeling and mathematical paradigm. The use of big data for intelligent development to drive system flexibility requires new challenges, such as how to ensure data consistency and confidentiality in long, complex supply chains. There are also many challenges in modeling that address how to properly model all of the key technologies, especially health, safety and the environment, and require accurate, predictable, but critical, quantities at specific locations. Environmental concerns require that we closely track all molecular species so that they can best be used to create sustainable solutions. The environmental impact of disruptive business models that stem from new, personalized products is unpredictable.