Utilising Lean Six Sigma to mitigate and manage COVID-19
The COVID-19 pandemic has resulted in societal disruption and loss of life at a global level. One of the most disruptive issues of the crisis is the uncertainty inherent to the virus and its management. The uncertainty in this context includes the economy, employment, finances, physical and mental health of people around us. Uncertainty can leave us feeling anxious, stressed and powerless over the direction of our lives. The professional discipline of Lean Six Sigma as a powerful Operational Excellence methodology has the potential to address some aspects of this uncertainty, enhance understanding and provide practitioners and academics in the field with effective guidance to mitigate and manage COVID-19.
Current epidemiological approaches towards COVID-19 are based on sound objectives: identify the cause and/or contributing factors, determine the extent of the virus, its progression in the global population, assess preventive and therapeutic measures and develop an effective public health policy response. The public health response by nation and state has varied with differing results, but one aspect of the response has involved testing various protocols. For tests to be most effective, a short turnaround time for test results is critical from a patient who carries this virus. Yet, the variation in turnaround has been wide, ranging from hours in India to days in the United Kingdom and other places. This undesirable variation in turnaround times can be quite fatal, especially with patients who have underlying health issues. A number of factors could potentially contribute to test turnaround time. Lean Six Sigma is a methodology which can be effective to seek out the contributing factors to this variation, diving deep into the components of the testing process from start to finish from a patient value perspective and determining the root causes of such undesirable variation using appropriate tools in the tool-box.
Lean is concerned with reducing all types of waste in processes whereas Six Sigma is about finding and eliminating causes of defects by focusing on process outputs, which are critical in the eyes of customers (Antony et al., 2018). The combination of Lean and Six Sigma reduces waste and improve consistency in performance through continuous reduction of variability in critical business processes and ultimately minimises operating costs, optimises productivity and maximises customer satisfaction. Testing for COVID-19 is essential not only for people themselves but also to isolate the people infected and trace and quarantine their contacts. In South Korea, a country with over 50 million people, 747653 people were tested until May 17th and 262 people have been declared deceased to date. Although the tests for COVID-19 are available in most countries today, times for testing can take from hours to days. Keeping variance as low as possible and eliminating root causes for the variation is key in an integrated approach (i.e., Lean Six Sigma) on mitigating and managing COVID-19.
Through a defined problem solving process, Lean practitioners increase value by eliminating waste in the form of activities that do not add patient value. With COVID-19, time getting tested for COVID-19 is of value, while waiting for those test results is not. Six Sigma was originally developed to demonstrate that quality is determined by reduction of process variation around the process mean and thereby achieving more consistency in the output performance such as waiting time for the present example. By creating challenging goals to reduce variation, Six Sigma practitioners utilise a prescriptive approach, called Define, Measure, Analyze, Improve, Control (DMAIC), to work together to find and understand the cause and effect relationships of factors that drive uncertainty (Harry, 1998). In the COVID-19 testing example, a LSS team applies the DMAIC methodology to the COVID-19 testing process, and uses a variety of statistical and non-statistical tools to understand the significant few factors, or causes, that drive the uncertainty of when test results show up for an individual.
The LSS approach aligns with DMAIC both strategically and tactically. LSS goals are strategic: If operations improvement is the goal, the DMAIC approach shows us how to get there. DMAIC is project based, with goals defining each phase to be met: Define opportunities, Measure performance, Analyze problems, Improve the situation, Control for sustainable performance. In most cases, there is no specification for turnaround times for the tests and one of the first steps is to understand the realistic specifications for turnaround times of tests through a benchmarking exercise and understanding how often we meet the specifications in the eyes of patients. With regard to COVID-19 testing, a DMAIC project could explore strategies to improve the testing process, both in terms of reducing the turnaround times using simple and effective tools of Lean such as Value Stream Mapping and how accurate the results are for patients using Six Sigma tools such as correlation, regression analysis and hypothesis tests. The following paragraph shows how DMAIC could potentially be applied in COVID-19 testing scenario.
In the Define phase, we need to define who the customers are and understand their expectations. Here one could utilise the use of Voice of the Patient analysis along with Kano Model to understand the expectations of patients and what might aid them. In the define phase, we need to understand the process through a simple process map from start to finish of the testing process. Moreover, SIPOC (Supplier-Input-Process-Output-Customer) can be very beneficial to have a high level understanding of the process. In the Measure phase we have to understand the current situation both from a process and data perspective. One should create a Value Stream Map (VSM) from start to finish of the testing process so we will be able to understand how non-value adding activities are delaying the testing process. Further, we have to create a plan to collect data for turnaround time for individual tests and determine the typical errors made during the testing process. In the present example, the critical-to-quality (CTQ) characteristic is turnaround time. It is important to establish acceptable specifications for the turnaround times through the Voice of the Customer (VOC) analysis. It is also paramount to establish a good measurement system in this phase so that accurate data can be collected for analysis. In the Analyse phase, we need to understand the potential causes of variation in turnaround times and also determine the errors and defect rate of the process. In the Improve phase, we need to develop potential solutions to not only reduce the mean turnaround times but also variation in turnaround times and reduce the number of errors which could lead to patient dissatisfaction. We also need to understand different pain points in the existing process and devise strategies to eliminate them in the eyes of patients. In the Control phase, our primary purpose is to monitor the improved process and make sure that changes we made to the process are statistically controlled. We should continuously monitor the newly established process and determine whether the processes are statistically controlled or not. It is essential to develop and document the new process and this will be the new standard for everyone to follow and sustain. All laboratory staff should be trained on the new process to ensure that no staff members will revert back to the old methods of testing.
Developments in information technology have the potential to improve the DMAIC approach further, taking advantage of the data revolution. This data revolution has been a period where the vast improvements in how data are produced are reaching into every endeavour. For example, “big data” are the massive volume of data that societies and businesses produce every day. With current technology, the ease by which organisations collect, and analyse these data, in real time, facilitates improved organisational decision making and predictive power. The huge volume of data generated daily on the pandemic could potentially help governments and healthcare providers take several actions. For example, Taiwan integrates the database of health insurance and customs to generate alerts during a clinical visit to assess an individual case, based on travel history and exhibition of any existing symptoms. As a result, a list of steps are created to track confirmed cases and to prevent any further infection or transmission. However, this example has escaped a problem of big data, namely, that most big data efforts fail, in part because many organisations have more data than they can effectively utilise. The Taiwanese case and other successful national examples are efforts of enormous costs and large national efforts. These are factors that cannot be replicated organisationally.
Yet, the media are reporting on type I and type II errors in testing, leading to false positive and false negative results, respectively. Successful testing must reflect a focused organisational effort. A focused LSS effort through DMAIC has the potential to help organisations meet the goal of an improved testing process. Organisations developing tests should organise a diverse set of subject matter experts around the goal of approved, accurate COVID-19 testing. An assessment of current testing performance capabilities should be performed using the vast historical data. Utilising specific Big Data tools, such as data mining, to identify, clean, and analyse data from testing capabilities. This action must be purposeful. A conventional big data process where warehoused data are collected en masse, but not accessed, used, or analysed for any organisational effort represents a wasted effort. A better approach is to use Six Sigma methodology to determine the most appropriate data and then use it to support strategic decision-making. Six Sigma methodology can also be used to support conventional clinical methods through hypothesis testing. A Six Sigma Big Data framework might conduct analysis and improvement through the deployment of machine learning. In analysing options for improved testing, machine learning assists individuals in identifying patterns to make smarter conclusions. Optimising for testing improvements and sustainability through a rigorous control of the process parameters in building an effective test may also utilise artificial intelligence and predictive analytics.
The testing differences presented above appear to be an ideal test case for big data solutions and Lean Six Sigma: a large, complex, but largely unstructured problem. A framework like DMAIC which is based upon rigorous attention to quality, involving subject matter experts, and organised within the domain of Lean Six Sigma can provide the structure needed. Combining these approaches has the potential to successfully address at least one area of COVID-19 uncertainty: the process of testing.
Authors
Dr Chad Laux, Associate Professor of Computer and Information Technology and Lean Six Sigma Black Belt, Purdue University, USA
Professor Jiju Antony, Professor of Quality management and Lean Six Sigma Master Black Belt, Heriot-Watt University, Edinburgh, Scotland, UK
Dr Gretchen A. Mosher, Associate Professor of Agricultural & Biosystems Engineering, Iowa State University, USA
Manal Alduraibi, PhD candidate, Purdue University, USA
Reham Nour, PhD candidate, Purdue University, USA
Willem Salentijn, PhD Scholar and Lean Six Sigma Master Black belt, Vrije Universiteit Amsterdam, Amsterdam, Netherlands
References
Antony, J. (2011), Six Sigma vs Lean, International Journal of Productivity and Performance Management, Vol.60, No.2., pp. 185-190.
Laux, C., & Springer, J. & Seliger, C. & Li, N. (2017). Impacting Big Data analytics in higher education through Six Sigma techniques. International Journal of Productivity and Performance Management, Vol.65, No.5, pp. 662-679.
Antony, J., Palsuk, P., Gupta, S., Mishra, D., & Barach, P. (2018), Six Sigma in healthcare: a systematic review of the literature, International Journal of Quality & Reliability Management, Vol.35, No.5, pp.175-1092.
Harry, M. J. (1998), Six Sigma: a breakthrough strategy for profitability, Quality Progress, Vol. 31, No.5, pp.60-64.
