Viewing SPC software as a “technical product,” headquarters takes a hands-off approach and has each plant choose its own software (often a hodge-podge of off-the-shelf software and inadequate spreadsheets). The result is a more costly Total Quality program in which SPC analysis is inconsistent. Determining and implementing process improvements that can be shared corporate-wide becomes confusing and costly.

The obvious answer to this problem is to standardize on one SPC software, which offers a variety of benefits:

Easier communication and comparison

Standardizing software makes communicating and comparing SPC information easier and less time-consuming for engineers and managers who need to monitor and analyze processes in multiple plants.

Take, for example, the case of a manager who oversees production of four plants, all making the same product. Each day he receives process reports from each plant. Two of the plants use the same SPC software, and the other two each use a different spreadsheet program. The manager’s staff spends hours each week reconfiguring information from the two plants that are using spreadsheets—just to get to the point where the analysis can take place. Those precious lost hours could be better spent on recommending improvements to reduce process variation and increase the bottom line.

Reliable decision-making

Given that vital manufacturing decisions are based on SPC analysis, it is surprising that more companies have not standardized their software. Using a consistent set of tools provides a much more reliable decision-making process.

If you are comparing control charts from four different plants, all created with different software, how can you be sure that plant A’s exceptional results are real and not due to the way it calculates its process capability indices? Standardized software would enable you to enforce that the Cpk, Cp, etc. are calculated the same way.

Faster results

One of the major complaints of organizations implementing SPC is that it takes too long to get to the point where they can make improvements because the first step must be getting the process in control. Once a process is in control and predictable, it can be improved through changes. Standardizing software speeds up the implementation process so improvements can be made sooner and gains from those improvements realized faster. Also, lessons learned in one facility often can be transferred faster to others because “the wheel” doesn’t have to be reinvented with each plant.

Easier personnel training

Standardizing SPC procedures and software greatly reduces training time for plant personnel. It also makes it easier to migrate procedures and technologies to new facilities. And internal user groups can emerge to provide guidance on common problems such as “what do you do for this process” questions. These peer groups also allow faster implementation of process improvements across multiple plants.

Justification for capital spending

When process improvements are quantified by believable SPC analysis, there is better justification for capital spending.

Consider the case of a multiplant chemical manufacturer where SPC analysis has determined that a new type of blender will reduce variation in a key characteristic’s constituent that makes up more than 60% of the cost. The improved process was quantified by the shift in mean of the control chart for this characteristic, which was then converted to a cost reduction value from which an ROI could be calculated. The new blenders required several million dollars worth of investment, but the ROI for purchasing the same equipment at the other plants was well within corporate goals.

IT department efficiencies

Standardizing software saves time for IT departments that are responsible for supporting applications used in multiple facilities. It is a simple fact that fewer applications mean the IT staff can focus their limited resources more deeply rather than more broadly.

Simplified quality procedures

International quality procedures such as ISO9000 and QS9000 can be simplified with one standard SPC tool. Standardizing would allow routine reports and analysis to be generated with fewer problems.

Supply chain consistencies

Multinational corporations with production facilities in diverse parts of the world expect to receive consistent product from a global supplier. Increasingly these multinationals insist on receiving SPC reports from suppliers, by each production plant, to demonstrate process control. Only then can the corporation feel comfortable that a supplier can handle its volume fluctuations and supply-side risks from multiple plants.

As manufacturers drive suppliers to deliver more consistent materials, the metrics for measurement can be compromised by inconsistent SPC analysis techniques. Standardizing SPC software greatly improves the likelihood that consistency goals will be reached.

More advantage from the Internet

The widespread use of corporate intranets for communication creates an even greater opportunity for comparing SPC analysis. A firm that is standardized on its SPC software can choose to make the analysis available to those who need it worldwide through the Internet. Recipients of this information could be internal plants, or the manufacturer's customers and suppliers. This is part of the evolution in supply chain management.

Maximizing large information systems

Given the worldwide trend for manufacturers to move to large information systems such as ERP, MES and SCADA, standardized SPC software can make an even bigger difference, allowing engineers and managers to extract further process improvements from the vast amounts of data these systems collect. Standardized SPC software can become an integral part of the overall system, playing the role of a corporate decision-support tool.

Let the users choose the SPC software

Standardization is only effective when you choose the right software in the first place. Some corporations relegate the evaluation and choice of SPC software to the corporate statistician, with the thinking that “it’s all just statistics.” This thinking is like asking a process engineer to predict next year’s orbit of Jupiter because it’s all just physics. 

The corporate statistician typically analyzes trends and forecasts ranges, which get factored into business plans and risk assessment. This person usually wants high-end, comprehensive, general statistics software.

Not so with SPC software users, who need a tool that can break down the process data into discriminating events, allow them to view them graphically from different perspectives, and give them a clear understanding of the causes of variation and probable actions to improve the process. General statistics software usually does not include the charting and analysis needed for these functions. It is also more complex.

The solution is to provide the right tool for each job: a general statistical package for the statisticians and dedicated SPC software for everyone else.

Going forward with the standardization

Uniting multiplant SPC software fiefdoms is the key to producing highly standardized products. Doing this doesn't have to be a painful process, but it takes commitment and drive to make it happen. The following tips have proven effective for companies who have succeeded with standardization.

1. Management at the highest possible level needs to decide to standardize and communicate that commitment throughout the company. This is the first and most important step, because without management's commitment, it's just too hard to get people to change what is familiar. The standardization should be presented in terms of a strategic decision to use a consistent set of tools for analyzing processes to make sure products meet specifications.
2. Set a reasonable timetable and appoint a champion who can get the job done. The champion should be someone who can convince others of the benefits of standardizing.
3. Choose SPC software that is flexible enough to cover the processes in each plant. Will it be easy to learn so novices can start, yet have automation features for more powerful applications? Make sure it will link with the manufacturing databases in your plants. Put the software through a “usability test” before deciding. Get feedback from users.
4. Devote adequate resources to implementation and training. Plan to develop multiplant user groups, if possible. User group members can be efficiently linked via Web sites or e-mail lists. 
5. Iron out the wrinkles by implementing in one or two plants first, then apply lessons learned to other plants. A successful implementation may also help pave the way with other plants because it proves it can be done effectively.
6. Drive the project through to completion.

Conclusion

With these kinds of benefits and the critical need for global product consistency, there simply is no justification for allowing SPC software fiefdoms to continue. Unite those fiefdoms, and you will be on your way to building an empire.

About the author

Clifford S. L. Yee is president of Northwest Analytical, Inc., maker of NWA Quality Analyst software. He holds an M.B.A. from Butler University, completed graduate studies in mechanical engineering from UCLA and holds a B.S. in aerospace engineering from the University of Michigan. He holds five U.S. Patents on engine cooling and is a recipient of the Center for Leadership Development’s Distinguished Award.

back to top