Finding the right fit for automation
The goal of achieving a fully autonomous factory that essentially runs by itself with little or no human involvement and zero waste, repair or energy loss is the future vision of most manufacturing organizations. Today, however, the industry is still in early phases of factory automation and manufacturers must take care not to apply technology for technology’s sake but to implement it where it makes most business sense, based on in-depth commercial and technical analysis.
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.
Between the ranges of full scale automation to no automation 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.
Evaluation criteria
When considering automation, manufacturing organizations should begin with the following steps:
Thorough audit of current manufacturing processes. A deep understanding and characterization of current manufacturing processes is paramount in order to deploy an optimized solution that will consistently meet both the technical and commercial targets of the business.
Strong business case. Organizations that apply automation in a reactive manner without justifying its value to the company often fall flat on their investment and the desire to leverage the technology within the company loses momentum. The overall objective, vision and strategy for automation must be mapped out and aligned to the digital transformation strategy of the enterprise in a formalized process, along with ways to best monetize its application.
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.
Design considerations
If an OEM requires automation as part of the manufacturing of a product, such as in the case of consumer medical products like pacemakers, then a contract manufacturer is required to automate the production line. However, it is still the OEM’s responsibility to ensure that the product design is suitable for automation. A design review must be conducted with a complete risk assessment and consideration of any commercial issues that could arise. If production is automated without a series of testing and validation to understand and address limitations, there will be technical gaps and a loss in product quality. When deciding whether or not to automate production, design engineers must take into account several factors, including:
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. If fittings or screws must be applied, sometimes this not only increases the part numbers but also the errors during production. Materials must be designed so that they are suitable for automated production.
Viability for automated inspection. 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.
Suitability for handling. 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 all together
As new automation technologies and approaches emerge in the manufacturing industry, there are many paths that an organization can take in the pursuit of maximizing operations, applying automation in the right places to capture value. When automation does make sense, the best approach is a carefully-phased deployment that includes strict review and vetting gates at every step to ensure optimal technological and commercial gains for the enterprise at large.
About the author:
Mulugeta Abtew is VP, Technology Development at Sanmina – www.sanmina.com
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
