实现一个完全自主工厂的目标，该工厂本质上几乎没有或没有人类参与，零浪费，维修或能源损失是大多数制造组织的未来愿景。但是，如今，该行业仍处于工厂自动化的早期阶段，制造商必须注意不要为了技术而应用技术，而要根据深入的商业和技术分析来实施它具有最大业务意义的地方。

For products such as automotive controllers and consumer medical products like heart monitors that require routine, pristine and error-free assembly, automation has become the defacto standard and a key requirement from original equipment manufacturers. For large telecommunications assemblies that are created to operate within network infrastructures, automation is not the optimal solution. These low volume products contain multiple heavy parts that require substantial investment in order to automate production, making the return unattractive. In addition, complex and large scale medical systems such as MRI machines are highly sophisticated products that require precision assembly with strict tolerances. Attempts to automate such assemblies demand costly and custom automation with limited scalability.

范围之间的全面自动化一个utomation across manufacturing operations, there are varying degrees of automation that can be applied to provide the best value at the right stage of a production process. For example, a single robot can be used to hand solder a product because it’s a redundant and simple step that provides a good return on investment to automate. Other examples include use of robotics for the lifting of large, heavy materials for products that are produced in very high volumes, or autonomous factory vehicles that can be summoned to deliver materials to the factory floor at the proper stage of a production process.

---

EvaluationC瑞典

考虑自动化时，制造组织应从以下步骤开始：

Thorough audit of current manufacturing processes。对当前制造过程的深入了解和表征至关重要，以部署优化的解决方案，该解决方案将始终如一地满足业务的技术和商业目标。

Strong business case。以反应方式应用自动化的组织而没有证明其对公司的价值的合理性，通常会依靠其投资和利用公司内部技术的愿望失去动力。必须将自动化的总体目标，愿景和策略映射到正式过程中，并与企业的数字化转型策略进行绘制，以及最佳利用其应用的方式。

Knowledge of current market offerings.To date, a large portion of data generated from automation is unusable. Some experts estimate that data scientists spend more than half of their time cleaning data. Organizations should look for automation systems with proven data collection capabilities that enable critical manufacturing data to be optimized for analytics, as well as predictive and preventative maintenance. Cloud-based manufacturing execution systems that aggregate, correlate and analyze data from machines on the factory floor provide the most flexibility and can be scaled up or down as needed.

---

设计Considerations

如果OEM需要自动化作为产品制造的一部分，例如在Pacemakers等消费医疗产品的情况下，则需要合同制造商来自动化生产线。但是，确保产品设计适合自动化仍然是OEM的责任。必须对可能出现的任何商业问题进行完整的风险评估和考虑，进行设计审查。如果生产是自动化的，没有一系列的测试和验证以了解和解决限制，则将存在技术差距和产品质量损失。在决定是否自动生产自动化时，设计工程师必须考虑几个因素，包括：

Number of product parts, assembly steps and other parameters。If production of a product is expected to be automated, the number of product parts and fasteners must be minimized, along with the corresponding number of assembly steps. Tolerances must be taken into account, as well as the inner spacing of parts. Even color and dimension are critical parameters that must be factored in when considering automation.

Product materials。如果必须使用配件或螺钉，则有时这不仅会增加零件数字，而且会增加生产过程中的错误。必须设计材料，以便它们适合自动生产。

自动检查的生存能力。A product must be designed in such a way that it can be automatically inspected. The challenge is to minimize false errors, such as rejecting a good product or accepting a bad one, and collecting the right data to predict and prevent risk or failures during the manufacturing of a product.

适合处理。The product must be capable of being handled, so that a robot or technician can pick it up and place it in another location as it goes through various production steps.

Bringing it一个二togher

随着制造业中新的自动化技术和方法的出现，组织可以采取许多途径，以追求最大化运营，在正确的位置应用自动化以捕获价值。当自动化确实有意义时，最好的方法是经过精心培训的部署，其中包括严格的审查和审查大门，以确保整个企业的最佳技术和商业收益。

关于作者：

Mulugeta Abtew is VP, Technology Development at Sanmina –www.sanmina.com