Today’s process control systems are capable of quickly handling enormous amounts of data in different formats and communicating with a wide variety of devices and equipment. While all of these features and technologies are allowing us to do things that may have never been imagined 20 years ago, what has been lost in the rat race is an emphasis on making it easy to set up, maintain, and operate these complex systems. After all, we have experts and trained specialists around to sort this stuff out . . . right?
Adding fuel to the fire is another important trend in our industry. Budgets are tight, and a large number of experienced workers (operators, maintenance technicians and engineers) are nearing retirement age. As they go from spending their days in the plant to dropping a worm in the local fishing hole, the knowledge that took years to accumulate, knowledge that is vital to operating and maintaining process plants and other facilities, is lost. The globalization of the world economy also results in plants being built in areas with little or no industrial history and therefore a limited pool of trained or even educated workers.
Human Centered Design
The solution to taming the complexity associated with today’s technology and creating products that plant personnel can quickly learn and use efficiently lies within an approach called Human Centered Design (HCD).
HCD is an interdisciplinary and iterative approach to design focused on understanding and meeting human needs versus an exclusive focus on features and technology. Let’s use the example of tax-preparation software, which facilitates quickly and easily filing your taxes. This is accomplished by designing the software in a manner that is intuitive with attention to the things that matter most like preventing errors and identifying tax credits. A typical filer would find it much easier to understand “Did you pay any interest on a mortgage?” as opposed to “Enter the value reported in box 1 of Form 1098 into line 10 Schedule A (Form 1040), subject to the limitations outlined in IRS Publication 936.”
When it comes to software user interfaces, this means using a task-based approach to better align the system with the user’s mental model about the work he or she is trying to accomplish. The concept of Human Centered Design is not new; its roots lie in Human Factors and Ergonomics research and design that arose during World War II to apply knowledge of human limitations and capabilities to the design of equipment, especially with aircraft cockpit controls to prevent crashes. The approach had been slow to reach the computer industry, as early users wore the challenge of memorizing arcane command codes and workarounds as a badge of honor. But as personal computers became more popular, and especially with the advent of the World Wide Web, companies became more aware of the need to deal with the “average user” who doesn’t have the time or inclination to study the programs and figure out what the programmer was thinking when they wrote the application. Rather, the forward-thinking companies found a competitive advantage in understanding the customer and designing the program to match their way of thinking.
Carnegie-Mellon University’s Human Computer Interaction Institute (HCII) pioneered research in this area and advanced the methodologies for applying Human Centered Design to computers and software user interfaces. They provided significant input to Emerson in developing its own Human Centered Design Institute. The Institute’s primary mission is to ensure each product the company introduces is aligned with customer work practices and goals. "Understanding our customer's work practices, how they interface with each other and technology, and performing usability testing, provides the foundation for Human Centered Design," says Institute director Duane Toavs.
The HCD Process
Applying Human Centered Design is not a matter of just gathering a bunch of engineers and applying common sense. For one thing, you will find that “common sense” is not that common (although once you come up with an innovative and intuitive design it should seem that way) and secondly this approach will result in an application that meets the engineer’s needs (often throwing in the latest cool technology fad) but makes no sense at all to the ultimate user of the product.
Applying Human Centered Design boils down to three key activities:
- Understanding the users. Understanding the users and their goals involves interviews, contextual inquiries (observing workers going about their daily work in their workplace), researching tasks and work standards, and analyzing current solutions. The first question to be answered is how the user thinks about their work and what makes them successful at the end of the day. This information is organized into typical user or “persona” profiles, scenarios, and task maps that show the important interactions between different users. Back to the tax example, the types of filers identified at the beginning of the article provide an oversimplified set of three personas (pay a professional to do everything, old school paper and pencil, and tax software consumer).
- Creating design solutions. With the user-centric information providing a strong foundation, design teams then embark on ideas for new products. The design phase includes sketching and ideation followed by a series of software or hardware prototypes.
- Testing the design. The evaluation phase follows, with design evaluations of the prototypes and usability testing with actual users. The whole process is iterative: design prototypes, evaluate with the aid of actual users, and then improve upon the design based on user feedback. Only after several of these iterations is the product actually ready to be built.
While these activities may seem obvious and easy to implement, they break from the traditional development approach where marketing writes a set of requirements and engineering builds something based on their best interpretation of the collective requirements. The evaluation phase and focus on iteration also present challenges to a traditional release schedule where the scope of a project is fairly well defined.