skip to main content

Factory Automation is Key to Sustainable Manufacturing

At Western Digital SanDisk Shanghai Assembly & Test Facility

By DIDIER CHAVET and SHEKAR KRISHNASWAMY

There has been an increasing emphasis in recent years on sustainable manufacturing, the term for production strategies that address traditional business and economic considerations but also seek to minimize negative environmental and social impacts.[1]

Factory automation is a key strategy used by many manufacturers to achieve more sustainable manufacturing because of its potential to reduce waste, inventory, unnecessary processing, and idle time. A pilot project at the SanDisk assembly and test facility in Shanghai, China, demonstrates the promise of this approach (see figure 1).


Figure 1. Material-handling activities at the Western Digital SanDisk Shanghai assembly and test facility are being automated in a move to achieve more sustainable manufacturing.

SanDisk, now a brand of Western Digital Corporation, is a global leader in NAND flash memory storage solutions that are at the heart of many of the world’s largest data centers, and are embedded in advanced smartphones, tablets and PCs. SanDisk Shanghai is a leading-edge assembly and test operation that produces fully assembled and tested flash memory modules. SanDisk also operates another facility in Kulim, Malaysia.

SanDisk Shanghai’s factory automation project is to automate material-handling activities using robotic vehicles and storage systems, and real-time factory dispatching and activity-management software.

SanDisk Shanghai is at the forefront of adapting to advanced manufacturing requirements with factory-automation strategies. The material-handling project is the latest in a series of automation projects at the plant (see figure 2) that have addressed shop-floor execution, quality, mobility and other functions over the past decade, and that will be used in big data and industrial Internet of Things (IoT) technologies in future automation systems.


Figure 2. The Western Digital SanDisk Shanghai assembly and test facility has become increasingly automated since a shop-floor execution system was deployed a decade ago. Big data and industrial IoT technologies are also in the process of being used in automation systems there.

Simplifying Complexity

Semiconductor manufacturing is inherently complex in many areas, such as management of human resources, increasingly challenging process requirements, dynamic equipment requirements, materials complexity and reliance on a complicated supply chain.

These challenges now also extend to semiconductor assembly and test facilities, because today’s more complex packaging technologies require such facilities to make greater use of wafer fab-like operations and strategies.

Inadequate management of these complexities drives waste (see figure 3), which in turn affects factory key performance indicators and, ultimately, the bottom line.


Figure 3. There are many sources and kinds of waste in semiconductor production facilities. Factory automation can reduce waste, thereby increasing sustainability and improving the bottom line.

Factory automation systems address these complexities because they enable optimum planning and utilization of resources, help minimize waste and lead to reduced capital and operating costs. This is critical to sustainable manufacturing and profitability because customer orders and priorities can change frequently, impacting factory-floor demands and manufacturing capacity. If manufacturing systems aren’t able to handle these shifting demands, then waste will be generated.

Reducing Errors and Waste

At the Shanghai facility, material-handling activities involved a number of manual and semi-automated processes that introduced opportunities for error, resulting in material waste, unnecessary processing and work-in-process (WIP) bottlenecks. In addition, the activity of product processing itself had become so complex it was not possible for human operators to comprehend all of the decision options and make the best decisions.

SanDisk decided to address these problems by using automation software. The specific goal of the project was to ensure that the materials needed for processing were automatically delivered to the right tool at the right time.

Applied Materials Real-Time Dispatch (RTD) and Activity Manager, both part of the Applied Materials Automation Productivity Family (APF) software suite, were selected to automate key steps in the manufacturing process. This included monitoring of vital production equipment prone to causing bottlenecks, such as die-attachment and wire-bonding tools.

The Applied RTD and Activity Manager systems provide two key capabilities. First is the ability to easily build sets of dispatch rules that enable optimum production throughput, and second is the ability to enforce those rules via a workflow-management system.

The result is a material-handling system that, by anticipating and reacting to factory-floor conditions automatically, enables optimum production. For example, if a given production tool needs to process a specific product lot at a certain time, the system will automatically know the location of that lot as a result of its previous process step. It will then dispatch an automated robotic vehicle (ARV) to retrieve the lot and place it on the tool. After processing is finished, the lot will automatically be taken to where it is needed next, or to storage in the warehouse. A schematic of the overall system is shown in figure 4.


Figure 4. The material-handling system illustrated above in the highly automated Western Digital SanDisk Shanghai assembly and test facility interfaces with the facility’s existing shop-floor execution and supervisory factory automation systems.

To date, the results from the SanDisk Shanghai project include successful lot-to-equipment allocation by Applied RTD, and initiation of lot transport tasks by the Applied Activity Manager system (using the dispatch results from RTD).

These project tasks, coupled with the physical and control systems of the automated guided/robotic vehicle (AGV/ARV) system, have enabled the company to reduce lot-to-lot variability because of more consistent and uniform manufacturing, and also to reduce operator errors and delays.

The next phase of the project will address inherent logistics bottlenecks and optimize the physical system to achieve higher levels of equipment utilization.

Conclusion

The SanDisk Shanghai project is demonstrating the usefulness of factory automation as a way to achieve waste reduction, promote more sustainable manufacturing and improve the bottom line of assembly and test operations.

Didier Chavet is APAC IT Silicon Operations Head at SanDisk, a Western Digital Brand, and Shekar Krishnaswamy is Business Development Manager at Applied Materials.

For additional information, contact didier.chavet@sandisk.com or shekar_krishnaswamy@amat.com

[1] http://www.mckinsey.com/business-functions/sustainability-and-resource-productivity/our-insights/sustainabilitys-strategic-worth-mckinsey-global-survey-results