Case Study: Streamlining Coast Guard’s Accounts Payable Process

 

The local senior leadership of the U.S. Coast Guard Finance Center was confronted by the problem of maintaining high levels of service in its accounts payable (AP) process in the face of increasing process complexity. Compounding the issue was the fact that the center’s information technology (IT) systems had been modified to accommodate the growing complexity without a comprehensive business process review, which in turn added even more complexity, resulting in a vicious and endless cycle.

The center, located in Chesapeake, Virginia, employed approximately 360 full-time federal employees and 180 contractors who provided a range of accounting transactional processing and financial statement preparation services for the Department of Homeland Security (DHS), Coast Guard and Transportation Security Administration (TSA). Beginning in October 2006, the center commenced full accounting services for the Domestic Nuclear Detection Office. It also operated and maintained the financial system and associated data bases for some components of DHS. The center processed approximately 2.5 million transactions annually.

Overcoming Process Complexity

An area of particular concern for the local senior leadership was the accounts payable process supporting TSA. It wanted this process to be world class and become the standard process for all their accounts payable services. The finance center leadership was convinced the process used the correct basic financial system tools and architecture for long-term sustainability, but the complexity of this process had grown since its inception of just two short years before. This complexity (Figure 1) over-stressed people, processes and systems with re-work loops, delays, errors, penalties, duplicate payments and more. The center was in a bind, and bringing in additional resources was not an acceptable option.

Figure 1: U.S. Coast Guard Finance Center Current-State Accounts Payable Value Stream Map

But the center’s senior leadership had a strategy for success: Bring in an experienced Lean Six Sigma consultant with a finance background to lead the project team, break through the complexity, and design a Lean, effective and scalable process.

Project Starts with a Value Stream Map

Where did the project team start? It began with a value stream map, then it identified where the non-value-added time was spent. Next, the team developed a measurement system so that it could determine how much time was spent in these non-value-added areas. In addition, the team began to measure queue volumes by specific activities over time within the process. For example, during the initial analysis, the team determined from the Define and Measure phases of a DMAIC (Define, Measure, Analysis, Improve, Control) project:

◉ Current process cycle efficiency is less than 1 percent

◉ Current lead time is 14 days

◉ Sigma level = 1

Some tough questions were asked and that forced many process owners and stakeholders to re-think why they were doing things the way they were. Finally, the project team measured the voice of the customer by conducting a phone survey with their TSA customers. Many Lean Six Sigma tools, such as process mapping, cause-and-effect analysis, failure mode and effects analysis, and basic statistical analysis, were used. In addition, two Kaizen events were conducted. This resulted in some quick wins.

One of the Kaizen events, in particular, produced tremendous results in the authorized certifying officer (ACO) invoice approval queue. This queue contains invoices that had already been entered into the system, and were approved by the contracting officer and his or her technical officer. The ACO invoice approval is the final step in the center’s payment approval process before the invoice is submitted to the U.S. Treasury for payment.

Fixing the ACO Queue Problem

This specific ACO queue was identified through the value stream mapping exercise as a constraint in the process due to the high level of items in the queue (an average of 175 invoices daily with spikes to nearly 700 at times). When the level of invoices spiked, overtime was required to process them. In one Kaizen event, code-named “Queue Blitz,” the ACOs did nothing other than review and approve invoices. After 10 days with intense focus, the queue reached an all time low of one invoice in the queue. That was almost a 100 percent reduction in work in progress (WIP) as shown in Figure 2.

Figure 2: Authorized Certifying Officer Queue

The next step was to maintain an acceptable level of WIP, and institutionalize this as a daily goal. The team determined that one day of WIP would equate to 80 invoices in the queue. (This was based on four ACOs reviewing and approving 20 invoices a day in addition to their other duties.) The finance center management teams are currently using control charts to manage the WIP to no more than 80 invoices a day with a stretch goal of 50. With this revised process, the level has never exceeded 83, resulting in little need for overtime and a reduction in interest and late penalties.

More Plans for the Future

This effort has begun to “turn the ship around,” but there is more ground to cover to get to a world class process. By utilizing the Lean Six Sigma methodology, the finance center now better understands where the pain lies (the root cause) and has already started developing improvement plans aimed at reducing the complexity and streamlining the larger process. The team identified process and management improvements that could be implemented quickly and expected to be completed with the Improve phase by the second quarter of fiscal 2007. After the process changes were implemented, the team defined areas requiring software changes or the use of new software tools, defining the requirements and performance outcomes for these software changes that will be given a high priority for completion and implementation

Continue reading

A Roadmap for Deploying Six Sigma in Small Businesses

Many Six Sigma experts have expressed doubt that Six Sigma can be used effectively in small, or even in some medium-sized, organizations. However, while the approach to deployment must be modified, it is possible for small businesses to successfully implement Six Sigma. Here is how.

At the outset, several givens must be in place:

◉ The owner of the business supports Six Sigma completely and actively, and is willing to personally spend time on it.

◉ The company has a routine core of work that will benefit from the process rigor of Six Sigma.

◉ The organization’s culture is open to change.

All businesses, but small businesses especially, must understand that Six Sigma is not a panacea. There are many aspects to business, and only some of them should involve Six Sigma. Business owners should not abandon their instincts, intuition, taste, feel for the market, competitive spirit, empathy with customers and employees, common sense or good judgment.

A business that wants to change must also meet three requirements:

1. Tolerance for variation and the failures that result. Change requires variation; in fact, it is defined by it.

2. “Slack,” i.e., spare resources that can be diverted to change-related activities.

3. Redundancy built into its systems, so that the areas being changed can still provide essential stakeholder services.

Big companies meet these requirements easily. But such is not the case with all small businesses. Their approach to Six Sigma must be modified in the following respects.

Increasing Tolerance for Variation

Small businesses cannot afford too many mistakes. Many are flirting with the line between survival and success. Thus, when choosing Six Sigma projects, the leader should err on the conservative side, especially in the beginning. Follow the rule used by successful professional gamblers: “Do not risk more than you can afford to lose.”

Before embarking on the Six Sigma journey, small businesses will want to be sure their customers are shielded from any problems that changes might cause. Prior to deployment, they must be sure that they have installed basic quality systems. ISO 9000 has proven to be useful in this regard. When conducting Six Sigma projects, small businesses must take special care to insulate customers from unintended consequences. Also, they must be prepared to forgive and forget when mistakes are made. The safe path of the status quo may result in fewer mistakes, but it is not viable in the long run.

Creating ‘Slack’ and Redundancy

Slack is the amount of time the change agents (Green Belts and Black Belts) spend on Six Sigma project work. Typically, a company’s most scarce resource is human talent. Six Sigma change agents must be a company’s best employees, so slack is the most important category and the most difficult to come by. Likewise, a company must have the ability to cover the important duties of these key individuals (redundancy.) The company leadership team must prepare a plan for creating this redundancy and slack before it launches its Six Sigma effort.

As a rule of thumb, a small business can begin deploying Six Sigma when it reaches a size where one person can devote one day per week to Six Sigma. Assuming an eight-hour day, this threshold is reached when total employment is 20 full-time equivalents (FTEs). This level of commitment is necessary to justify the time and money that must be spent training the change agent, educating the leadership and orienting employees.

This falls in line with what should be a small company’s maximum change agent commitment – 0.5 to 1 percent of its total employee hours (i.e., 20 employees [800 hours a week] means devoting no more than eight hours a week of one employee’s time to Six Sigma projects). The total time spent on change will be much greater than just the change agent’s, and will include time for team meetings and time spent by others implementing the changes. Too much change all at once can be disruptive to normal operations. The 1 percent rule will keep things manageable.

For companies with fewer than 100 employees, Green Belts should be added when total employment reaches 20 and 40 employees, rather than increasing the workload on a single individual. This is recommended for several reasons:

◉ It is usually easier to create small amounts of slack in different areas than it is to replace 40 percent of a key person’s time.

◉ It will create Six Sigma expertise in more areas of the company, which will help create a culture where Six Sigma can thrive.

◉ It will be easier to work on cross-functional projects if there are trained people in more areas of the company.

◉ It will more quickly create a change-agent community where people can learn from one another and share a common bond.

The company should consider rotating people through the Green Belt position, which will require additional training expenditures. Of course, a cost/benefit analysis should be conducted before investing in training additional people, but by then the company should have seen the benefits of Six Sigma and be willing to reinvest some of its gains to spread Six Sigma through the organization.

Growing Six Sigma

A company should stay with two or three active Green Belts – rotating Six Sigma projects among them – until the company reaches a size of about 140 employees. At this size it is large enough to have a full-time change agent, i.e., a Black Belt. Assuming that a company has two active Green Belts, its commitment to change will be 1 percent after the company hires its first Black Belt. (That is calculated as 5,600 hours x 1 percent = 56 hours, or one full-time Black Belt and two part-time Green Belts.)

It is not a good idea to replace all of the Green Belts with a Black Belt when the minimum for the 1 percent rule is reached at 100 employees. Black Belts do better when there are Green Belts with whom they can work.

Ideally the Black Belt should be chosen from the ranks of the company’s Green Belts, unless they are uninterested or clearly unqualified for the greater technical skills required of a Black Belt. An individual who not only has successfully completed Green Belt projects, but who exhibits a passion for the role is a good choice. The company will need to invest in Black Belt training, of course. Costs vary widely; a training program should be chosen on the basis of operational requirements as well as cost. If a Black Belt from outside the company must be hired, a knowledgeable consultant can help assess candidates.

As a company grows, its investment in Six Sigma process improvement projects should grow at a rate of one additional Black Belt and two additional Green Belts for every additional 140 employees. While a company has fewer than five or six Black Belts, the Black Belts should report to local supervisors.

However, when the company reaches approximately 700 to 840 employees, it should consider creating a formal Six Sigma organization headed by a full-time Six Sigma Champion. This individual should possess strong leadership skills and should report to the CEO. Black Belts are more effective when they report to a central Six Sigma organization. Typically, Black Belt success rates, measured by the value of completed projects and Black Belts who complete certification requirements, are about twice as high when Black Belts report to a Six Sigma Champion instead of a local or functional leader. There are two reasons for this. First, centrally reporting Black Belts are in a better position to work cross-functional projects. And second, local leaders often cannot resist the urge to have talented Black Belts work on their current local priorities, which, while important, are less urgent than Six Sigma projects.

Other Challenges

In addition to insuring a tolerance for variation and failure, and creating slack and redundancy, small businesses face additional obstacles not encountered by larger organizations. Two of the more daunting challenges are lack of expertise and the especially dynamic nature of small business. Here are some suggestions for dealing with these problem areas.

Inadequate Expertise – Use leverage to create “Super” Green Belts and to provide additional Black Belt support:

◉ Invest in software and books on Six Sigma. This “expert in a box” approach is dangerous in the hands of amateurs or Six Sigma newbies, but it is a necessary risk.

◉ Get help from large customer companies and suppliers. Caution: Tap into their expertise, but be wary of getting bogged down in their bureaucracy. Some Six Sigma programs have become remarkably hidebound.

◉ Local college faculty often can help with statistical expertise. Caveats:

     ◉ Six Sigma is not academic research. Know when to cut the analysis and act.

     ◉ The KISS rule (keep it simple, stupid) applies. Be sure the faculty member uses the simplest approach possible. Choose a person who can explain things in layman’s terms. (A good test might be to ask the candidate to explain binary logistic regression.)

     ◉ The faculty member may not understand what Six Sigma is. The company’s Green Belts and Black Belts might need to guide him or her.

◉ Use semi-retired experts. After 20-plus years of Six Sigma, there are plenty of people around who understand it and have used it. Find them.

◉ Take advantage of free support: iSixSigma.com’s online discussion forums, articles and information; the International Society of Six Sigma Professionals (ISSSP), etc.

◉ Cut travel costs by using online training and consulting.

◉ Hire interns from local colleges or universities. Juniors, seniors or graduate students can provide a lot of help with number crunching, data gathering, preparation and cleansing, and many other time-consuming tasks.

◉ Commission projects to be done by college students. Students are frequently assigned projects by their professors, and they are looking for partners. Be one.

◉ Many individuals are working to be certified as Black Belts and have passed a subject matter exam, but they need successful projects. Small companies can provide project opportunities.

It is important that outsiders sign non-disclosure agreements before being given access to proprietary information. This requirement might need to be bent for professors at research universities.

Dynamic Nature of Small Business – Most Six Sigma projects take four to six months, which is often too long in a small business environment. However, long cycle times often are the result of big company bureaucracy. They are not a built-in limitation of Six Sigma. Choose projects carefully, sponsor them effectively and pursue them aggressively. A small business will find that it can successfully complete most projects in four to six weeks, instead of months.

Continue reading

Lean Six Sigma for Poets

Lean can be of great value in office environments. However, the use of complex jargon and statistics, plus a focus on manufacturing, have hindered the adoption of these tools in other settings where they can be useful.

I was a liberal arts major in college while almost all of my friends majored in hard-core subjects like nuclear engineering, material science or biochemistry. (I believe they did this because wanted a thing called a “job” when they graduated.) One of the textbooks I had from my freshman physics class was called Physics for Poets. Now you can image my friends’ delight in tormenting me when they got a look at the dust cover of this masterpiece. They spent weeks thinking up new ways to insult me for not taking “real” classes.

However, I was pretty much mock-proof and their insults were lost on me. The reason was that I didn’t really give a rat’s behind about their version of physics. I couldn’t care less about the differential equations they used to prove the laws of thermodynamics, the atomic weight of barium or whatever arcane thing they studied. I only had an interest in the practical applications of physics for someone like me who was never going to be a scientist.

I craved pragmatic knowledge, like did the oil in my car really break down from engine heat every 3,000 miles and need to be changed? Or could I store bottles of vodka in my freezer? I was interested in real-world applications that would help me in my everyday life. I have a similar interest in the tools offered in Lean. (By the way, the answers to the two physics questions above are no and yes.)

My first exposure to Lean occurred twenty or so years ago when the general manager of the large manufacturing business I worked in purchased a book on the Toyota Production System for each member of his leadership team. Several weeks later in our team meeting he was talking with our production manager and said to her “That’s exactly the kind of muda we have to address right now.”

Now, I’ve never been known for my good common sense or for the ability to keep my mouth shut when I really should, so I asked him to repeat what he just said. He repeated it and I then asked what the word muda meant. He said it’s the Japanese word for waste discussed in the book he gave us several weeks ago. Being very quick on my feet I said, “Oh, I didn’t get to that part yet.” My team member friends burst into laughter. It turns a discussion of muda was in the book’s introduction. This began my twenty years of a tortured love-hate relationship with Lean.

I’ve been fortunate to work in a number of highly technical and process-oriented research and manufacturing companies. These firms put a premium on process improvement and high-quality products. The firms were sufficiently successful to offer abundant opportunities to learn and grow your personal capabilities. One of the areas in which I have received significant training and exposure is in the area of Lean and Six Sigma. The training I’ve had had been conducted by experts in the field with much experience and education. I’ve had the chance to take the training and in multiple formats over many years. Most of the training I have had I really hated.

I should perhaps state at this point that I love to learn and try new things; I am not a total luddite. I’ve got an advanced degree from a premier party school; I’ve taken subsequent classes in law, participated in executive retreats at Ivy League schools and read most anything I can get my mouse to click on. I’m good at math and love statistics. My master’s thesis used statistical modeling to attempt to predict job attachment. However, Lean training and I, we’ve had our issues. These issues I believe have hurt the ability of regular working folks like me to utilize Lean’s excellent tools to their potential.

My main concern with Lean training I have had is the cult-like focus on using Japanese words to describe everything. It seems to me that if you really want to impart knowledge to me you would go out of your way to make the tools easy to understand. I have asked probably a half-dozen expert instructors why the tools and techniques are described in Japanese and have heard a number of well thought-out explanations, which never made sense to me. It feels to me like Lean is presented almost like some kind of secret society of which only a chosen few can really enjoy the benefits. However, no one has ever accused me of being the brightest bulb in the chandelier, so let’s just assume there is a good reason which I am not clever enough to understand. Still, I think the language hurts the ability of regular working people to use the tools.

The other concern I have with the Lean training I have had is that it is exclusively focused on manufacturing process improvement, and the generic training is often focused on vehicle manufacturing. This is completely understandable and appropriate as that is where the most common use occurs, but I have had trouble seeing myself as an office person with office processes in the training I’ve had.

That said, although I’ve struggled with the training, I have loved the concepts and have worked over the years to apply the tools, frequently with English terms, to the work I have been engaged in.

Continue reading

Optimize the Total Costs of Quality

Boosting a product or service offering with optimized costs is a strategic objective of most manufacturing and service industries in the world. The ways to achieve this objective can range from shortsighted organizational downsizing and payroll cuts to farsighted plans to combat hidden costs. In the latter tactics, management exerts tight control over cost of quality (COQ) to achieve more with less while balancing the trilogy of profitability, customer satisfaction and employee satisfaction. If inflated, COQ becomes a serious silent killer that eats profitability for breakfast.

Monitoring and controlling COQ is indispensable to survival.

What Is COQ and COPQ?

COQ and cost of poor quality (COPQ) are sometimes erroneously thought to be synonymous, but COPQ is one component of COQ. COQ, sometimes referred to as total COQ, is the sum total of costs associated with preventing failures and appraising quality level and costs resulting from failures. So COQ is made up of two main components:

1. Cost of good quality (COGQ) represented by prevention and appraisal costs, and

2. COPQ for failures costs.

Failure costs are divided into external costs (supply chain costs) and internal costs (field failure costs). Total COQ can be represented in the equation below.

Figure 1: Equation for Total COQ

COQ Components

The key differentiator between internal and external failure costs is whether they occur before or after reaching the customer. Internal failures are those resulting from a product or service’s nonconformance to requirements that occurs before reaching the customer – whether those failures take place in any of the design, procurement or production processes. For instance, if a finished product requires rework due to design changes, it takes place before delivery to the customer and is an internal failure. Similarly, replacing defective raw materials acquired from a supplier happens during the procurement process and is considered an internal failure.

Figure 2: Examples of Internal Failure Costs

Conversely, external failure costs are those costs incurred by a product or service’s nonconformance to requirements that occurs after reaching the customer. Toyota’s 2019 recall of vehicles due to unintended acceleration is a poignant example of an external failure, which caused 52 deaths, 38 injuries and a financial loss of $5.5 billion. Another example of an external failure is the tragic 1986 Space Shuttle Challenger explosion that occurred 73 seconds after takeoff, causing seven deaths and a financial loss of more than $1 billion. A more mundane example would be when a customer returns a defective product to the seller for a refund or makes a claim against the manufacturer’s warranty.

Figure 3: Examples of External Failure Costs

On the other end of the spectrum exists the COGQ that, if leveraged, combats poor quality costs. Despite being costs, appraisal and prevention activities are desirable to a certain extent beyond which the law of diminishing returns dominates. Raw materials being inspected (appraisal activity), for instance, will decrease the odds of having nonconforming input to the manufacturing process. Even better, reviewing and rating your suppliers regularly (prevention activity) increases the odds of receiving consistent quality materials which could eliminate the need for frequent inspections down the road.

Although appraisal costs are considered COGQ, they must be used sparingly. This is because they are detective rather than preventive activities. For example, inspection of a finished product at the end of the line would probably detect defects, but will never eliminate the root causes; thus, defects recur. Appraisal costs are costs incurred to determine the degree of conformance to quality requirements – not to prevent causes of failure. This component of the COQ can take place anywhere during procuring raw materials, producing an item, or could be in activities external to the organization such as inspections, tests or audits conducted at the site for installation or delivery.

Figure 4: Examples of Appraisal Costs

Prevention costs are the other part of the COGQ; they are the costs of all activities designed to prevent poor quality in products or services. It is best to maximize the equation as they keep failure and appraisal costs to a minimum. Cost to eliminate a failure after delivery is five times that at the development or manufacturing phase. Prevention activities, therefore, are best performed at upstream rather than downstream processes. For instance, reviewing a design before release to manufacturing would spare the organization extra costs that might be incurred due to internal or external failures after production. Prevention activities might be deployed anywhere in the value stream, from marketing through production.

Figure 5: Examples of Prevention Costs

Optimizing TCOQ

Going back to the total COQ formula mentioned above, logically yet still controversial, the TCOQ value cannot be zero. It is an optimization problem. Since nobody is operating in a perfect world, no organization can ever produce or offer a defect-free product or service without deploying appraisal or prevention measures. Obviously, COGQ components – appraisal and prevention costs – need to be maximized, while COPQ components – internal and external failure costs – are to be minimized to reach to the minimum TCOQ value.

Figure 6: Optimizing COQ

As shown in Figure 6 above, COPQ declines as the quality level improves, but this doesn’t occur without exerting some level of prevention efforts. The key question that the TCOQ formula should answer is: To what extent should the organization invest in COGQ so that it reaches the minimum TCOQ with the optimum quality level of the product or service? The answer points to where the organization needs to position itself consistently to retain competitiveness or, more bluntly, to survive.

Improving a product or service quality level while keeping profitability at decent levels is not a walk in the park endeavor. It means having a proper grasp of the COQ concept and methodology, coupled with a standardized approach of monitoring and controlling over optimum levels of cost. Such mastery will allow the organization to balance the competing demands of profit and customer and employee satisfaction.

Continue reading

Tips for a Successful Virtual Gemba Walk

As we work remotely throughout this great crisis, the need for Lean and continuous improvement persists. COVID-19 continues to eke its way through the last months of the 2020 and uncertainty still looms large. But continuous improvement professionals ought to be lights in their organizations. The continuous improvement paradigm should propel practitioners to continue exhibiting the fundamentals of executions that will empower our organizations to come out of 2020 stronger than before. To Lean Six Sigma professionals, there is no “waiting it out” and no permissible “wait and see” approach. Improvement is a slope that we’re either moving up or down, improvement or stagnation.

It is important, therefore, to hold fast to the foundational disciplines with which our discipline is built. Standardization, visual management and respect for people will be key differentiators for our departments and organizations. The gemba walk is at the intersection of these three lynchpins. There are, however, lots of unknowns about conducting a gemba walk in the virtual space.

A gemba walk is an in-person walk through a defined process to assess the current state. The foremost purpose of the gemba walk is people-engagement. When walking a process in the office, hospital or factory floor, we ought to be engaging others to think critically about the current state. Everything else – whether it’s 5S, checklists, kanban or process flow – is secondary to exemplifying respect for people in your walk.

The following are some suggestions for conducting a gemba walk in the virtual space.

◉ One on one: Don’t try to conduct a virtual gemba walk with five people in a Webex call. This will lead to low participation and group think. Just as before, having one-on-one conversations will deliver the greatest current-state insights.

◉ Use cameras: Virtual meetings are opportunities for distraction and multitaskers, but that’s what there is for the time. By using cameras in your virtual gemba walks, you can curb some of these natural tendencies to become distracted. By being on camera yourself, you demonstrate the importance of engagement from your team.

◉ Set a cadence: Make sure you communicate to your team when you’ll be doing your gemba walk, which process you’ll be walking and who you’ll be walking with. Setting a constancy of communication will deliver accountability and build trust between you and your team.

◉ Screen share: If it’s applicable to the process, screen share may allow you to see parts of the process that you couldn’t see before. The virtual space provides new opportunities to look at parts of your process that may have seen too obtrusive back in the office. Make the most of this opportunity!

◉ Business as usual: Aside from these best practices, try to keep as much of the previous structure as possible. Keep the same performance metrics and don’t deviate from historic talking points, at least to start. Too many changes can overwhelm the team and lead to lower engagement in your Lean journey.

In order to be true champions of continuous improvement, we cannot wait out the crises facing our organizations. We must find ways to adapt and improve even in the most challenging of circumstances. To this end, the gemba walk is still foundational. The virtual space is an opportunity to engage with your team in a new and refreshing way. By leveraging the technologies and best practices discussed, you can ensure you keep engagement for continuous improvement activities high.

Continue reading

Case Study: Improving Purchase Order Process

The accounts payable shared services center (SSC) of a large Ohio-based healthcare system completed a Six Sigma operational enhancement and reporting project. This case study reviews the course of the project which endeavored to establish standard invoice processing times and develop an ongoing system for monitoring how well those standards are met.

Over the last three years, the SSC processed an average of 738,000 invoices a year for payment. The largest segment of its operations was, and continues to be, purchase order (PO)-based invoices. In 2013, PO invoices accounted for 56 percent of the total processing volume for the SSC.

As a means to help address the high volume of PO invoices being processed, the SSC established a daily production quota. The apportionment was determined by taking the estimated total number of PO invoices to be processed in a given month divided by the number of working days within that month. This number was then divided by the number of assigned employees. The formula used to calculate the quota was a frequent complaint of SSC employees and supervisory staff because it failed to recognize the variability in processing time when multiple line items were present on a PO invoice.

Project Objectives

Amid calls for change, the SSC embarked on a Six Sigma project in mid-2013 that contained three objectives.

1. Establish a standard time range for completion based upon the total number of line items on a PO invoice.

2. Determine the process time for an invoice and whether that time fell below, within or above an established standard time range (also based upon the number of line items on each PO invoice).

3. Develop a comprehensive operations report that displayed employee processing times measured against standard time ranges.

Database Evaluation

Within days of the project’s kick-off, the systems analyst for the SSC evaluated the accounts payable application used with PO invoices. The findings revealed the application’s database design was unable to separate PO invoices by total number of line items or to capture the total processing time expended by the employees who worked on the invoices. Discussions between the SSC project team and the developers of the application led to the conclusion that the best long-term solution was to make modifications to the database. Outside developers were contracted, and the database was modified to pull in all necessary data and make it readily accessible.

Data Analysis

Fourth quarter 2013 data was exported from the database into Microsoft Excel by structured query language (SQL) code written into Microsoft Access. The exported data, referred to as raw data by the project team, was presented in three columns. These columns contained the name of the employee, the total number of line items, and total time, in seconds, for each PO invoice processed. The project team copied the content of the columns into Minitab to perform data analysis. Despite being confident that the transactional raw data possessed a non-normal distribution, the project team performed a graphical summary on the response variable, processing time.

Figure 1: Summary of Processing Time, Q4 2013

The Anderson-Darling statistic, with its p-value being lower than the chosen significance level of 0.05, confirmed the assumption that the data did not follow a normal distribution. This cued the project team to evaluate various nonparametric tests and determine which of these tests were the most appropriate going forward. An important factor in this decision was the wide variance in processing times. The graphical summary displayed a range of 3 seconds to 3,875 seconds. The project team needed a nonparametric test that was effective against outliers and errors in data. After a thorough review, the decision was made to use the Mood’s Median Test.

With the nonparametric test selection now behind them, the project team focused its attention on acquiring an accurate understanding of the effect the total number of line items had on actual processing time of a PO invoice. Again using the fourth quarter 2013 raw data, the project team carried out a Mood’s Median Test. The outcome of the test revealed that the 33,567 PO invoices consisted of 77 separate groups of invoices possessing the same number of line items. Further, the project team found that 33,924, or 98 percent, of the PO invoices encompassed the first ten invoice groups.

Figure 2: Number of Line Items Per PO Invoice

Due to the high percentage of invoices in those first ten invoice groups, SSC management decided that standard time ranges would include only those first ten groups. The project team next needed a statistical tool to help identify the standard range of processing times specific to each invoice group. Because of the ability to provide an upper and lower bound, the confidence interval was the instrument of choice. Obtaining a useful confidence interval within Minitab’s available nonparametric tools, however, gave the project team an additional challenge. While the Mood’s Median Test produced the required median information for each invoice group, the corresponding confidence interval failed to provide a usable range as it does not include lower and upper bounds.

Figure 3: Standard Processing Times – 1

The solution was Minitab’s bootstrap macro. The macro calculates nonparametric confidence intervals by using bootstrap methods code, which is a statistical technique used for making certain kinds of statistical inferences and involves a relatively simple procedure repeated so many times that the technique is heavily dependent upon computer calculations. To apply the macro, the project team had to identify input data, the number of iterations to be used in estimating a confidence interval and the significance level. With processing time as the input data, iteration volume at 1,000 and a significance level of 0.05, the bootstrap macro was performed on the first invoice group – PO invoices with one line item.

Figure 4: Bootstrap Macro – 1

With 24,923 data points, the bootstrap macro was unable to provide any separation between the upper and lower bounds. To address this issue, the project team had to reduce the volume of data points being calculated. This was accomplished in a two-step process:

1. Calculate an appropriate sample size. The data point figure of 24,923, substituting as a population number, was input into an online sample size calculator. The result was a sample size of 378.

2. Randomly select and segregate 378 of the first invoice group’s processing times. To achieve this function Microsoft’s Random Sampler was employed. The project team subsequently copied the 378 processing times into Minitab and re-performed the bootstrap macro.

Figure 5: Bootstrap Macro – 2

The bootstrap macro was utilized on the remaining nine invoice groups. To further broaden the distance of upper and lower bounds for invoice groups 2 through 4, the project team again used the same two-step process.

Figure 6: Standard Processing Times – 2

At the end of this analysis, a standard processing time range for each distinct invoice group had been successfully constructed. This meant that the project team had realized the first of the project’s three goals (establish a standard time range for completion based upon a PO invoice’s total number of line items).

The remaining two objectives – the ability to determine an employee’s elapsed processing time and whether that time fell below, within or above an established standard time range and the development of an all-inclusive operations report – were determined to be mutually dependent. This conclusion guided the project team to seek added assistance from the systems analyst.

Developing a Time Analysis Report

As discussions began, the project team laid out the requirements for a time analysis report. The contents of the report needed to include the standard time ranges by invoice group, including median numbers, names of employees, median processing time of each employee by invoice group, and a color scheme applied to an employee’s processing time when compared to the standard time range. In addition, the project team wanted the user of the report to be able to designate a date range and either run the report or export the raw data into Microsoft Excel. The conversations that followed contained debate over content, format, coding, application strengths and weaknesses, and completion date. In the end, agreements were reached and the development of the report started.

In order to meet the needs of the time analysis report, the systems analyst built not only the report but also a command menu page to accommodate the functionality. An extended review of the relationships between the deliverables influenced the decision to first build the command menu page. Using SQL and visual basic for applications (VBA) coding within Microsoft Access and SQL server software, the command menu page was built.

Figure 7: Command Menu Page

To create the time analysis report, the systems analyst used the same coding and software tools used to produce the command menu page. Due to the report’s complexity, however, a custom module using VBA coding was developed and a conditional report formatting instituted. Lastly, links between the command menu page and the time analysis report were created.

With the time analysis report fully functional, the project team ran the report to acquire fourth quarter 2013 baseline performance figures.

Pilot

Figure 8: Baseline Figures

The project team met with the SSC employees to present the report and baseline numbers. A four-week pilot tested the internal theory that employees whose performance results are openly and routinely shared among members of their work group (as an application of a Lean visual office) strive harder to meet or exceed current operational standards. The feedback from employees was predictable – those employees meeting or exceeding standards had no concerns with the sharing of information while those who found themselves failing to meet the standards were not entirely enthusiastic about the idea. What ultimately sold the concept to all associates was that everyone’s performance was being shared. Additionally, improved employee efficiency would be reflected in the raw data when standard processing time ranges were recalculated. This would then generate a positive downward trend in processing times. Throughout the duration of the pilot, a time analysis report was produced weekly and distributed by email.

Results

While pilot outcomes showed a measurable improvement in productivity, the pilot revealed that employee motivation was not solely responsible for the gained efficiencies. Employees who found themselves struggling to meet the standard processing time ranges sought and received coaching from supervisory staff. The extent to which the supplemental instruction influenced the conclusion of the pilot could not be determined. An unexpected discovery was the fact that the employees lacked a knowledge of the tasks required to correctly process PO invoices. Regardless, the pilot was jointly deemed an overall success by the project team and management.

Following the pilot and its disclosures, the SSC assembled and conducted a series of employee retraining sessions aimed specifically to address trouble areas. Operational policies were updated to prevent possible processing time manipulation. New standard processing time ranges, based upon pilot data, were set in place. Employees continued to receive the time analysis report weekly and their capacity to meet PO invoice processing requirements was tied to bi-annual and annual appraisals. On the whole, the SSC creatively overcame every obstacle encountered and intends to aggressively employ this successful methodology on all future projects.

Continue reading

Case Study: Reducing Delays in the Cardiac Cath Lab

Cardiac catherization labs represent a significant capital investment for many hospitals. Realizing a return on this investment is increasingly challenging, given the introduction of advanced technologies and limitations in reimbursement. To meet the challenges and maintain fiscal health, hospitals pursue Six Sigma, Lean and change management techniques to improve throughput, maximize equipment utilization and increase efficiency.

New York-Presbyterian Hospital embarked on a comprehensive initiative aimed at improving throughput in the cardiac catherization labs at the Columbia University Medical Center, New York Weill Cornell Medical Center and Children’s Hospital of New York-Presbyterian sites.

The improvement initiatives at New York-Presbyterian focused on the various sub-cycle times impacting throughput – including case start time, room turnaround time and patient prep time. As a result of these multiple projects, the hospital gained 312 hours of procedure time without incurring any additional capital expense. An overview of one project conducted at Children’s Hospital of New York demonstrates how the Six Sigma DMAIC (Define, Measure, Analyze, Improve, Control) methodology provided the framework and tools to raise departmental productivity by improving first case start times.

The Define Phase

Improving first case start time was selected as a project by the Children’s Hospital of New York for several reasons. It contributed to a significant amount of lost productivity and failure of the first case to start on time was delaying subsequently scheduled cases. This variability in start time and lack of schedule predictability also was contributing to staff, physician and patient dissatisfaction.

A charter was developed and approved by senior leadership and a team assembled to lead the initiative. The charter provided:

• Project Scope – This established the parameters for the project. The start point of the cycle was patient’s arrival at the hospital and the end point of the patient’s entrance into the cath lab. The charter also described areas outside of the team’s scope, such as room turnaround time, which was the focus of another team.
• Business Case and Problem Statement – Baseline data indicated that 62 percent of the first cases were not starting on time representing 267 hours of lost staff productivity and unused procedure capacity annually.
• Goal Statement – A goal of 80 percent on-time starts was established.
• Team Members – The team for the project included the cath lab director, staff, cardiologists and anesthesiologists. The vice president of operations served as project sponsor and oversaw the work of the team.
• Timeline – A timeline including frequency of meetings, dates and times was agreed upon at the team’s first meeting and proved essential to keeping the project on track.

The project charter provided the team with focus and direction. The team then developed a map describing the current process.

Figure 1: High-Level First Case Start Process Map (Source: GE Healthcare and New York-Presbyterian Hospital)

Completion of the process flow map pointed to one opportunity for immediate improvement – streamlining the nursing assessment. One of the department’s nurses routinely calls patients the night before to reinforce pre-procedure instructions. Discussion during the process flow mapping exercise revealed some redundancy in the information gathered during this phone call and the nursing assessment completed the day of the procedure. The team agreed that initiating the nursing assessment during this phone call would eliminate duplicate data collection, and shorten the time needed to complete the assessment the day of the procedure.

The Measure Phase

The team used brainstorming and a fishbone diagram to identify all the potential contributors to delaying the start of the first case. Some of the factors identified included:

◉ Patient arriving on time

◉ Registration process

◉ Transportation

◉ Timeliness of patient prep

◉ Completion of assessments by the cardiologist, anesthesiologist and nursing

Figure 2: Brainstorming and Prioritizing Critical Xs (Source: GE Healthcare and New York-Presbyterian Hospital)

Data was then collected to identify those factors having the most significant impact on delaying the start of the first case.

The Analyze Phase

Regression analysis, a statistical tool used to model and predict the relationship between variables, revealed that the time in which the cardiology assessment was completed was a key driver in whether the first case would be completed on time. The R-sq adjusted value showed that it accounted for about 60 percent of the variation in the process. Here is a table showing the first case start data’s statistical analysis:

X	Test	Results	Statistically Significant?
Nurse	Test for Equal Variances	p=.725	No
Nurse	Moods Median	p=.583	No
Nurse	Regression	p=.762	No
Latest Assessment Time	Moods Median	p=.432	No
Latest Assessment Time	Test for Equal Variances	p=.132	No
Latest Assessment Time	Regression	p=.177	No
Anesthesia Yes/No	Moods Median	p=.710	No
Anesthesia Yes/No	Test for Equal Variances	p=.318	No
Oral Pre-Med Yes/No	Test for Equal Variances	p=.981	No
Oral Pre-Med Yes/No	Moods Median	p=.288	No
Anesthesiologist	Moods Median	p=.389	No
Anesthesiologist	Test for Equal Variances	p=.013	Yes
Anesthesiologist	Regression	p=.625	No
Patient Arrival	Test for Equal Variances	p=.909	No
Patient Arrival	Moods Median	p=.615	No
Difference vs. Card Assessment	Regression	p=.042	Yes
Time Patient on Table vs. Card Assessment	Regression	p=0.00	Yes
Difference vs. Anesthesia Yes/No	Regression	p=.532	No
Difference vs. Nursing Assessment	Regression	p=.658	No

Source: GE Healthcare and New York-Presbyterian Hospital

The Improve Phase

The team used this information to discuss and develop plans to ensure the cardiology assessment could be completed in a timelier manner. For example, since the cardiology fellow typically initiates the cardiology assessment, the director of cardiology explored other responsibilities and obligations that might be interfering with timely completion of the assessment. As part of developing a revised process, the team also completed a new process flow map indicating a target completion time for each step in the process that ultimately would lead to the desired case start time.

As shown in the table below, re-measurement of the process indicated a dramatic improvement in the number of first cases starting on time and a reduction in variation.

Data Categories	Baseline Data	Improve/Control Data
On-Time First Case Start	38 Percent	83 Percent
Baseline Z	1.44	2.47
Median	13 Minutes	0 Minutes
Mean	38.24 Minutes	6.33 Minutes
Standard Deviation	55.62 Minutes	22.4 Minutes

The Control Phase

Process control mechanisms were implemented to ensure the changes could be sustained, and that the gains achieved from improvement activities would not be lost over time. The control plan outlined the procedure for monitoring the critical X (completion of cardiology assessment) as well as the number of on-time first case starts. Regular reporting to the project’s executive sponsor reinforced the importance of the initiative and insured that changes would become imbedded into the organization’s culture.

Continue reading

High-performance Teams: Understanding Team Cohesiveness

Teams are the basic structure of how projects, activities and tasks are being organized and managed within companies worldwide. Global organizations striving for competitive advantage are increasingly incorporating the use of high-performance teams to deploy complex business strategies.

Work done in teams provides many advantages and benefits. The major advantages are the diversity of knowledge, ideas and tools contributed by team members, and the camaraderie among members. A characteristic commonly seen in high-performance teams is cohesiveness, a measure of the attraction of the group to its members (and the resistance to leaving it). Those in highly cohesive teams will be more cooperative and effective in achieving the goals they set for themselves. Lack of cohesion within a team working environment is certain to affect team performance due to unnecessary stress and tension among coworkers. Therefore, cohesion in the work place could, in the long run, signify the rise or demise of the success of a company.

Stages of Team Development

Team development takes time and frequently follows recognizable stages as the team journeys from being a group of strangers to becoming a united team with a common goal. According to researcher Bruce Tuckman, in both group dynamics and the four stages of team development he popularized (forming, storming, norming, performing), leaders must retain the motivation of team members in order to successfully overcome the challenges of the storming and norming stages (Figure 1).

Figure 1: Stages of Team Development

The forming stage represents the beginning, the honeymoon period; great expectations are shared from all team members. Relationships are developed, purpose is clear and ground rules are established. The storming stage is triggered once team members start jostling for position, stumbling from confusion, having arguments about leadership, strategy and goals. This is when team leadership becomes imperative. The leader must succeed at keeping the team motivated, addressing all concerns and clarifying purpose and goals.

Once the storming stage is overcome the team is ready to establish open communications, stable positions and norms – the norming phase. Trust is finally gained, and “when the trust account is high, communication is easy, instant, and effective.” These are the first steps towards cohesiveness. Once cohesiveness is achieved, teams will move from norming to performing and subsequently to highly performing.

What Is Cohesiveness?

Cohesiveness is the extent to which team members stick together and remain united in the pursuit of a common goal. A team is said to be in a state of cohesion when its members possess bonds linking them to one another and to the team as a whole.

Members of a highly cohesive team focus on the process, not the person; they respect everyone on the team, assuming good motives; and they fully commit to team decisions and strategies, creating accountability among the team. Morale is also higher in cohesive teams because of increased team member communication, friendly team environment, loyalty and team member contribution in the decision-making process.

Successful business strategies are usually carried out by an effective team with a high level of team cohesiveness. Highly cohesive teams are more committed to the goals and activities, are happy when the team succeeds and feel part of something significant, all of which increases self-esteem which in turn increases performance (Figure 2).

Figure 2: Cohesiveness to Performance

Key Drivers to Achieve Team Cohesiveness

High-performance teams are what make companies successful. Whether the task is to create an innovative product or service, or to design a new process or system, teams rather than individuals are assuming more of the load than ever before. The ideal team combines individual talents and skills into one super-performing-whole with capabilities that surpass those of even its most talented member.

High-functioning teams are not the result of coincidence. They achieve greater levels of participation and collaboration because their members trust one another, share a strong sense of team identity, and have confidence in their abilities and effectiveness. Such teams possess high levels of team emotional intelligence (EI).

Figure 3: A Look at Emotional Intelligence

EI is generally defined as encompassing the awareness and understanding of emotions (Figure 3). It incorporates the application of this understanding to decision making, regulation, and self-management: these three are all important aspects of teamwork. Studies have shown that EI has a positive impact on teamwork by making the team more cohesive.

Building an emotionally intelligent team requires developing emotional competence for the group as a whole. Teams that enjoy high levels of EI have established norms that strengthen trust, group identity, and efficiency. As a result, their members cooperate more fully with one another and join forces more creatively in furthering the team’s work.

Research from organizational behavior experts Vanessa Druskat and Stephen B. Wolff suggests following three practices to build your team’s EI:

1. Make time for team members to appreciate each other’s skills.

Interpersonal empathy is critical to confidence. Interpersonal empathy builds confidence within team members; once team confidence is achieved, individual confidence will then follow. The team must be aware of each member’s skills and personality. People on teams in which they knew one another better were more efficient and got more work done. When team members know the individuals they are working with, they attain a different level of trust with each other. Trust increases motivation and motivation increases commitment; once the two exist within a team you have achieved cohesiveness, which in turn increases performance.

Team-building activities are a great way to implement this EI-building practice. For example, when a team is formed at one company, organize several team-building activities – even outside of normal work hours and location – so that the team members get to know each other better and develop empathy. Planning meetings outside of the workplace builds camaraderie. Team-building activities also reinforce organizational commitment as the team perceives that the company cares about the success of the team as a whole. Games are a way to engage team members and learn about each other’s skills on a more personal level.

2. Raise and manage emotional concerns that can help or encumber the team’s progress.

It is important to establish comfortable, team-endorsed ways to express the unavoidable anger, tension and frustration that arise in a team effort and to positively redirect that energy. Both humor and playfulness can be helpful tools in resolving conflict and relieving tension and stress.

A couple of examples come from two well-known companies.

◉ At innovation consultancy IDEO, team members tossed soft toys over cubicle walls when feelings ran high. Besides lightening the mood, this action served as a reminder that the group had established norms for expressing difficult emotions, thereby making them feel less threatening to individuals and to the group as a whole.

◉ In another approach, Xerox team members wrote down their gripes, clipped them to play money in denominations from $1 to $100 depending on how serious they felt the issue to be, and dropped them into an “opportunities” jar.

3. Celebrate success.

Building the EI of a team also requires the expression of positive emotions such as gratitude and admiration when exceeding expectations. Recognizing individual and team achievements not only fortifies a team’s identity, but it also spotlights its effectiveness and collective passion for excellence.

Team lunches after a good project has been completed is an example of a typical celebration. The whole team goes out to a restaurant to celebrate their success and hard work. Another example is featuring a team’s accomplishment in a monthly newsletter from the CEO, so the entire company learns what has been accomplished.

Above and Beyond Cohesiveness

Although cohesiveness is a crucial and determinant factor for team effectiveness, cohesiveness alone will not guarantee success without organizational commitment. Team members can feel cohesion with their teammates but be completely detached from organizational values and vice versa.

Highly effective teams must have both perceived team support (PTS) and perceived organizational support (POS), but their PTS must be higher than their POS. PTS can be defined as the “degree to which employees believe that the team values their contribution and cares for their well-being.” POS can be defined as “the extent to which employees believe that the organization values their contribution and cares about their well-being.”

Above and beyond cohesiveness there are other subset factors that are important in team development to make teams work, as shown below and in Figure 4:

1. A clear set of objectives, communicated explicitly by management

2. Metrics allowing team members to evaluate their performance and the connection between the work of the team and key business indicators

3. Ongoing training

4. Decision-making authority necessary to reach business goals

5. Team-based rewards and appraisal, not only individual incentives

6. An open culture with easy access to relevant information and to senior management as needed

Figure 4: Factors Involved in Team Development

Final Thoughts

Teams drive organizational success, though developing and leading high-performance teams is one of the most complex tasks facing any leader in the current competitive work environment. Cohesiveness is the key factor in implementing effective, high-performance teams. Emotional intelligence also plays a key role in building high-performance teams in that emotional intelligence fosters cohesiveness. Managing emotions is how you build a team, an organization. It is the ability to get team members inspired. Leaders must understand how team cohesiveness works and how bonding in a team will build energy. Leaders must inspire team members through reinforcing the sense of belonging, empathy in bonding and mutual respect, in addition to giving people choice and power over what they can do. Once that sense of support, that foundation, is created, the result is limitless creativity.

Continue reading