The Sawhney Model: Operational Excellence for the People, by the People
The Sawhney Model: Operational Excellence for the People, by the People
ISE Magazine January 2021 Volume: 53 Number: 1
By Rupy Sawhney, Ninad Pradhan, Enrique Macias de Anda and Carla Arbogast
The question, “What are the signs operational excellence has succeeded in an organization?” has elicited a variety of responses over the past 40 years. To some, it is an improvement in quality and customer satisfaction. To others, it is a reduction in waste and variability. To yet another group of practitioners, it is sustainability and enhanced productivity. The philosophy of operational excellence, in all its ﬂavors, has undoubtedly inﬂuenced the way we approach the art and science of industrial engineering. But is there a better way to phrase this question.
Drawbacks to the traditional operational excellence mindset
Evidence of the burden operational excellence strategies can place on people has long been hiding in plain sight. Take the example of lean, one of the great success stories in operational excellence in the past few decades. Industry and academic experts underscore the importance of “getting it right,” to avoid the clear and present risk of implementation fatigue or failure. The numbers tell a story of sustainability percentages that get smaller each passing year after implementation, forcing organizations to introspect what wrongs could be righted the next time around.
Consider the growing momentum of Industry 4.0 and its redoubtable promise of better utilization of data and resources in the pursuit of improved productivity. Technologies to support Industry 4.0 continue to mature, yet at the highest levels of inquiry, such as the National Science Foundation and its Future of Work program, efforts to understand the impact of the impending technological disruption on workforce skill acquisition and mental preparedness are being encouraged.
The fabric that binds an operational excellence strategy to the workforce is often frayed. This is not surprising when we consider operational excellence from the viewpoint of the people who implement it and are most impacted by it. When organizations think about improving productivity, indeed when entire countries think about improving productivity, they consider an upward tick in the number of working hours as the ﬁrst positive sign of change. Yet, data from the World Bank show that the most productive countries usually work a lower number of hours than their lesser productive counterparts.
Regardless, people designing operational excellence strategies assign more tasks or more working hours to themselves and their peers. This is the widespread notion of operational excellence in the U.S., a country in which most employees experience fatigue at work, as the National Safety Council points out in its 2017 report. Is this where we want to lead a society already coping with increasing challenges in mental health and opioid dependence?
Perhaps it is time for an industrial engineering practitioner to recognize that productivity and quality of life must go hand in hand. The question that must be asked is: What are the signs that operational excellence has improved employee quality of life in an organization?
Toward people-centric operational excellence
This was the question that my team and I began to investigate about a decade ago at the University of Tennessee. This investigation quickly led to my reaching a fork on the road as a researcher in operational excellence. One path led to the pursuit of tactical goals and incremental contributions to well-studied topics such as lean tools, work design and applied optimization. In this pursuit, the inclusion of people-speciﬁc factors such as culture, stress, engagement and motivation would be considered tangential or incidental to the success of a project.
The other path led to pursuing strategic goals that would challenge our team to deﬁne transformational actions in an organization. We would place employee quality of life in the spotlight and, along with productivity, make it an inalienable criterion for success in an operational excellence framework. It was clear to us which path aligned more closely with our philosophy of operational excellence.
Nevertheless, any process of creating a new operational excellence model requires translating ideas into strategy and strategy into actionable targets. We based our journey on two industrial engineering pillars.
The ﬁrst is systems thinking. Systems thinking supports the visualization of the complex, qualitative relationships that exist between quality of life, productivity and sustainability. We formulated the requirements of the model by making these connections and by understanding their tradeoffs and feedback loops.
The second is critical problem-solving. This is the centerpiece to translating the strategic requirements of the model into tactical milestones. Concepts from widely used problem-solving approaches such as DMAIC and DRIVES, the latter developed by our team, are employed in the problem-solving pieces of the model. Beyond this, we consider our model to be evolutionary and dynamic, evolving to reﬂect the state of the art for speciﬁc techniques.
The proper name under which the model is published: “A Conceptual People-Centric Framework for Sustainable Operational Excellence.” Though colloquially and for the purposes of dissemination, we simply call it “The Sawhney Model.”
The Sawhney Model principles
We establish four principles for the Sawhney Model. First, an organization must reduce the required resource and effort by strategically deﬁning the problem. Second, all efforts must align clearly with measures of system growth and competitiveness. Third, systems must reliably enhance throughput and capacity. Fourth, transformations must sustain by enhancing employee quality of life. These principles encapsulate the systems thinking and critical problem-solving mindset.
We translate the principles of the model into a template comprising four modules. Module 1 identiﬁes the most relevant problem to be solved in an organization based on an analysis of the critical path that constrains system growth. This reduces the time and resources spent in pursuing tangential goals that are often the source of additional work and stress to employees.
Module 2 determines the performance metrics (leading indicators) that are instrumental to outcome metrics (lagging indicators) in an organization. The outcome metrics are organized such that they directly connect operational measures (e.g., capacity, throughput) to societal measures (e.g., employee quality of life, reputation).
Module 3 develops solutions based on improving the lagging indicators by making processes more reliable from the perspective of their constituent resources: People, material, equipment and information. The focus toward reliability and away from the traditional focus on efﬁciency empowers a “servant-leader” philosophy, in which management endeavors to create a conducive work environment for employees.
Module 4 anticipates the sources of employee resistance by estimating the impact of solutions on employee quality of life and mitigates them by making appropriate modiﬁcations to work design. This gives a deservedly high priority to factors that make an organization truly “people-centric,” such as culture, motivation, engagement and work-life balance.
The Sawhney Model is the basis of instruction for several undergraduate and graduate courses and educational programs at the University of Tennessee. The people-centric perspective has attracted graduate students to our research team, some of whose interest in the model is inﬂuenced by their corporate experience in operational excellence.
Guilherme Zuccolotto, formerly a lean Six Sigma black belt practitioner and now a doctoral student in the group, says, “I came from Brazil for a month-long summer program organized by the Sawhney group, was attracted to the ideas and practice of the model, and decided to return a few months later to pursue my doctoral research in people-centric operational excellence.”
Any operational excellence strategy is effective in achieving its goals only if the practitioners and employees consider it so. So how do these stakeholders perceive the model? We present this using two anecdotal experiences.
From homes to cars: The Sawhney Model in industry
The Sawhney Model experienced many technical modiﬁcations since its inception, but a “stable state” was reached around 2017, with the identiﬁcation of the four modules from Figure 2. We created a living laboratory arrangement for validating the model in which our industry partners would beneﬁt from the outcomes of the model whereas our group beneﬁted from access to the real-world challenges of its planning and implementation.
One of our earliest partners in implementation was the Clayton Homes facility located in Rutledge, Tennessee, a builder of manufactured housing and modular homes. “We faced high attrition rates at the Rutledge facility at a time when we wanted to increase production,” said Marty Mansﬁeld, general manager of the facility. Prefabricated home production is a labor-intensive line of work and we wanted to know: Could we alleviate some of that intensity?
We equipped every worker on the line with activity trackers and discovered that some workers walked an astounding 10 to 11 miles per day to perform tasks, switch between tasks and retrieve materials for the line. We surveyed workers and discovered that their work fatigue levels were high. We analyzed video and observed that tasks were shareable and some could be concurrent. This evidence-gathering process is typical of Modules 1 and 2 of the model.
It is pertinent to note that we consider questions like, “What is the step count per person?” and “What is the stress level per person?” on par with typical industrial engineering analyses such as “What is the cycle time?” and “What are the scheduling opportunities?” Elevating people-centered metrics to the same level of importance as production metrics allows us to innovate accordingly in Modules 3 and 4 (developing and sustaining solutions).
The solutions for Clayton Homes are industrial engineering solutions but with a distinct identity that can only be ascribed to the philosophy and organization of our model. Consider this list of novel methods designed and validated for the facility: a work-sharing-based scheduling algorithm to reduce cycle times; ergonomic workload balancing to reduce effort; and just-in-time supermarkets to simplify material availability.
According to Mansﬁeld, “We piloted the approach on one station and were surprised to see that it motivated people on other stations to come up with their own versions of it. The lack of employee resistance to the radical change was remark-able.” Clayton Homes has seen a dramatic rise in retention rates, which informally can trace connections to the Sawhney Model.
The second story takes us to a different work environment altogether. A day in the life of an automotive supplier facility is quite distinct from that of a house manufacturer. The cycle times change from tens of minutes to tens of seconds. We partnered with DENSO, a subsidiary of Toyota Motor Co. and a leading supplier of advanced automotive technology, systems and components for major automakers that operates its largest U.S. manufacturing facility in Tennessee.
The company approached us with the challenge of improving disruption recovery times in one of its manufacturing cells. The cell comprised six “zones” occupied by one employee each, the person being responsible for one or multiple steps in each zone. We interpreted this as a people-centered work design problem. That is, could we deﬁne a set of simple policies for the employees on the line to improve the recovery time of the line following a disruption? This reduces the emphasis on immediacy and urgency in response to disruptions, thereby addressing a common job stressor.
Consistent with Module 2 of the Sawhney Model, we focused on analyzing the zones leading to and following a disruptive event. The concept of ﬂoating bottlenecks, in this case, was found to be appropriate to capture the variation and disruptions to the production ﬂow in the cell. We collected 64 hours of data by direct observation and 40 additional hours of movement data using “indoor GPS” technology. Simulation models were generated and showed that guiding the reaction of employees to disruption were key to resolving the issue. More concretely, we found that the opportunity resided in deciding the timing of production stoppages following a disruption at some location in the cell. This counters the intuition and indeed the practice of “stop all work immediately” following a disruption.
Our simulation models suggest a stoppage policy that while an immediate stop in the ﬁrst and fourth zone would enhance shift throughput by 43% and 10%, respectively, immediately stopping work in the ﬁfth zone would decrease throughput by 16%. This insight helped us propose a no-cost disruption recovery strategy for DENSO that demonstrably improved the throughput while minimizing employee stress by equipping workers with simple disruption response policies.
“The focus on simple, low-cost and low-stress policies for our line workers introduced us to a new perspective on improving throughput while still taking care of the stress on our employees,” said Dan Dougherty from DENSO.
These are just two of many examples of applying the Sawhney Model in collaboration with our partners. We have ﬁve active projects in which we translate the principles of the model into practice.
We anticipate that the next frontier of the Sawhney Model resides in integrating it with information-centric industry frameworks. Industry 4.0 is an example of a framework used in industry to connect systems and services and to build a corpus of data. How do we adapt our model to work with millions of data points, thousands of employees sharing work and hundreds of sensors broadcasting data concurrently?
The critical observation is that companies must continue to focus on problem deﬁnition and selection, alignment of metrics with organizational goals, enhancement of reliability in the system and engagement with people. In other words, Industry 4.0 does not disrupt the core objectives of the Sawhney Model but does pose technical challenges in its implementation.
There are two challenges to resolve. The ﬁrst challenge is to enhance individual techniques in the Sawhney Model to keep pace with the pervasiveness of data; its collection and storage methods must factor into our implementation strategy. A highly connected system also implies that a small change in one subsystem can lead to big effects in a different subsystem. We must identify techniques that can cope with this complexity; for example, apply machine learning models to make sense of the interrelationships in the data.
The second challenge is to maintain the integrity of the model by continuing to be sensitive to the needs of the workforce. In fact, this challenge may present an opportunity to employ technology to bolster worker training, for example, using augmented reality. Another opportunity is to provide decision-makers with a live snapshot of the state of the system and recommend improvements, analogous to the concept of digital twin and virtual manufacturing that have gained momentum recently.
This is the history, state and future of the Sawhney Model. It has been a memorable experience translating our ideas into a people-centric operational excellence model and motivating to ﬁnd champions in industry and federal agencies who have chosen to support our journey. But what drives us is what lies ahead.