Creating a more flexible manufacturing line: one company changed successfully from producing a few high-volume products to greater quantities of more diverse products.

Traditionally, our customers, who are consumers of electronic ballasts for lighting systems, have been primarily interested in reliable, cost-effective products. The best way for our company to meet that demand had been with dedicated, hard-tooled automation, optimized for large production runs.

Recent national concerns about energy shortages, combined with an increasing awareness of the need for energy efficient solutions and systems to meet environmental measures, have changed the nature of the products end users want. Today, high-efficiency lighting controls, such as ballasts, are evolving quickly, and the range of applications for these devices is expanding. This evolution has put a demand for flexibility on our production that it was not designed to provide.

At our electronic control systems division, we focus on new and emerging technologies to provide energy efficient lighting solutions for applications ranging from the needs of offices, retail stores, hotels and warehouses to street lighting. Our business model for this facility has us moving away from very large runs of products, using dedicated automation, to one of smaller runs of many more models, suited to a broader range of lighting solutions.

To meet these new market demands, we needed to rethink our entire manufacturing process, primarily to enhance flexibility while maintaining the cost-effectiveness of our production facilities. In short, we had to offer more, without seeing our manufacturing costs skyrocket.

What's Wrong with the Old Ways?

In the past, we placed odd-form components, such as box capacitors, large electrolytic capacitors, connectors and transformers, with dedicated automation. This equipment has historically been built to satisfy the high-volume, low-mix manufacturing model. Although it worked well for us for a long time, it suffers from a couple of drawbacks: Because of the tooling, production changeovers are relatively time consuming and inefficient; and the assembly stations require dedicated technical support. Our evolving production requirements turned these drawbacks into serious limitations. Our new requirement was for equipment that could make production changeovers quickly and easily, while still managing the odd-form placement tasks with accuracy.

Defining What You Need

Before we started the search for new manufacturing equipment, we needed to know exactly where we were going. We wrote a detailed technical specification that defined exactly how we wanted the machine to operate, including our requirements for throughput, equipment uptime and insertion reliability--requirements we knew we could actually measure accurately over a defined number of parts run and within a certain period of time.

Due to the nature of our product, we had to ensure that after the transition we would have in place highly automated through-hole and surface-mount production processes that included flexible odd-form placement. Additionally, we required equipment that would integrate easily into our existing production lines, both physically and in terms of data collection and control.

Although the need for increased manufacturing flexibility and odd-form placement drove our search for new manufacturing equipment, rising labor costs were also a factor. As long as we would be upgrading our production, we determined to select equipment that would reduce the headcount required to build our products. Hand insertion is expensive and slow; to be competitive, we needed to take advantage of as much automation as possible.

Armed with a daunting wish list, we began the search.

Making Comparisons

When we began running benchmarks, we selected equipment from a variety of different vendors that met our basic specifications and offered two particular characteristics important to us: machine vision and an active clinch mechanism. To make accurate comparisons, we sent our people to the vendor's facilities with sample printed circuit boards (PCBs) and the parts we wanted to insert.

We already knew that our biggest insertion challenge came from the variety of parts that had to be placed on the boards (Figure 1)--in particular, the line needed to handle transformers with 0.025-in. (0.6 mm) square wire leads, which we knew were difficult to automatically insert properly. And, although equipment specifications are an important part of the research you do in selecting equipment, we needed to know beyond a shadow of a doubt that the equipment we chose would either insert all components and their leads properly or reject bad components. In essence, we required that the vendors actually show us that their equipment met our requirements, not just point to what their sales literature and brochures said.

[FIGURE 1 OMITTED]

Another of our requirements was that the equipment be SMEMA compatible, which would help ease the installation process. SMEMA compatibility means the new equipment can communicate with existing equipment both upstream and downstream, allowing us to drop the machine in-line with no changes to our product, its components of the board layout. This requirement would guarantee us a rapid and virtually seamless transition to the new equipment.

A final issue that we determined would be critical to the success of a manufacturing upgrade was the quality and quantity of support we could expect from the equipment vendor as we migrated from the old equipment to the new. Getting accurate answers quickly is critical. Any delay can be costly. We had to feel that the vendor we chose understood our manufacturing problems and would be willing to help us final appropriate solutions and not just talk about how wonderful their equipment is.

And the Winner is ...

Ultimately, the system that we picked met all of our requirements--it can place 1,440 components per hour (2.5 tact time) with the vision inspection on or 1,800 components per hour (2.0 tact time) without vision inspection. We found the system easy to program, and it has a wide range of feeders that provide the flexibility we need. And, from the beginning, the vendor, Universal Instruments Corp. (Binghamton, NY), worked with us to maximize the flexibility of the equipment--the Polaris Assembly Cell--so that the system would easily place all of our odd-form parts.

The Results Are What Count

The new equipment gave us a bonus we had not counted on. With a 47.25-in. (1.2 m) platform footprint, it actually has a smaller footprint than some of the dedicated automation equipment we were seeking to replace, which met our requirement for physically fitting into our existing production lines (Figure 2).

[FIGURE 2 OMITTED]

We did have to make one compromise: We slightly modified our product, increasing the diameter of the holes for the connector pins. Some of our products have connectors with as many as 10 in-line pins. A higher risk of a bad insertion of one of the connector pins exists than with any other component lead on the board.

The installation went well. Our technicians learned to program the machine while at the vendor's facility, and we received additional on-site support during installation and commissioning that included the installation, programming assistance and additional operator training on the machine. The support staff stayed with us until we were satisfied with the quality, reliability, up time and throughput of the equipment. In short, the support staff ensured that the actual installation reflected the specifications that we had benchmarked.

Ken Kindness is director of operations at OSRAM SYLVANIA's Electronic Control Systems Division (Lake Zurich, IL); e-mail: Ken.Kindness@sylvania. com.