case study 5
Making the Case for Quality
Clean Approach Saves Global Pharmaceutical Drug Manufacturer Time, Money
• A team-based Lean Six Sigma project aimed to reduce equipment cleaning time.
• Using a variety of quality tools, including process mapping, brainstorming, and root cause analysis, the team created a new cleaning procedure.
• Meeting all objectives, the project resulted in significant annual cost savings.
• The team was named a finalist in ASQ’s International Team Excellence Award Process.
At a Glance . . .
The cleaning procedure at Mallinckrodt Pharmaceuticals’ salts facility in St. Louis, MO, was hindered by significant bottlenecks. The fill, boil, and drain method, used to clean the five 2,000-gallon produc- tion tanks, was highly inefficient, resulting in rework and reducing capacity. In fact, production in this building had been on back order for more than a year; with an average changeover time between products of nearly three days, and a clean-out failure rate over 50 percent. Company leaders set out to identify ways to improve the efficiency of the process.
About Mallinckrodt Pharmaceuticals
Mallinckrodt is a global specialty pharmaceutical business that develops, manufactures, markets, and distributes specialty pharmaceutical products and medical imaging agents. The company’s Specialty Pharmaceuticals segment includes branded and specialty generic drugs and active pharmaceutical ingredients, and the Global Medical Imaging segment includes contrast media and nuclear imag- ing agents. Mallinckrodt has approximately 5,500 employees worldwide and commercial presence in roughly 70 countries. The company’s fiscal 2013 revenue totaled $2.2 billion.
Selecting the Project
To find impactful improvement projects, Mallinckrodt uses a rigorous selection process to evaluate which potential initiatives are most directly linked to the company’s strategic goals. Table 1 shows the various tools used in the selection process.
The proposed project focused on procedures for cleaning equipment after one product was made and before employees could start manu- facturing a different product. The process was so time consuming and ineffective that it was viewed as low-hanging fruit. “Any improve- ment in the process would shorten the timeline and allow additional time for manufacture of actual product,” said lead validation engineer Cindy Duhigg, “which is profit rather than time wasted.”
by Janet Jacobsen
April 2014
ASQ www.asq.org Page 1 of 5
Table 1 – Data and quality tools play key roles in project selection at Mallinckrodt.
Tools/data Why this is used
Project hopper To ensure that projects are aligned with the company’s goals and objectives.
Gemba + To take management to the front lines looking for waste and opportunities.
Brainstorming + To provide a wide range of ideas.
SWOT analysis + To evaluate each project’s strength, weaknesses/limitations, opportunities, and threats.
Affinity diagrams + To organize large sets of ideas produced during the brainstorming sessions.
Takt To determine the time required per unit output.
Feasibility study To provide a historical background of the project, description of the product, accounting statements, details of the operation, financial data, and legal requirements to estimate the project’s chances of success.
Project rating To ensure the project has a suitable cost-benefit ratio.
• Mohamed Razouk, operational excellence leader • Josh Steele, manufacturing engineer technician IV • Robyn Patrick, chemical technician III • Bill Gast, process engineer • Shirley Gause, lead operator • Amy Slovacek, industrial engineer • Josh Blough, production supervisor • Athena Tanner, manufacturing engineer technician IV • Bob Mohr, manufacturing maintenance
Using Quality Tools to Improve the Process
The first step in developing an effective improvement strategy involved pinpointing issues in the current process. Some of the quality tools used to accomplish this task are highlighted in Table 2. Of these tools, root cause analysis identified key drivers for the first-time right issues and revealed a 62 percent cleanout failure rate the team needed to address.
Next, team members narrowed the list of potential improvements by analyzing data related to the current state. At this stage, process maps revealed several cycle-time issues within the batch records, and these issues were prioritized in a cause-and-effect matrix. Of the possible
ASQ www.asq.org Page 2 of 5
The primary factors that supported this project’s selec- tion included the following:
• It was aligned with company strategy. • The project would improve the health of the
business with cost savings through energy and water conservation.
• It would afford the opportunity to increase production capacity and reduce back orders.
• The implementation of modern technology would yield a higher-quality cleaning process.
• It showed potential to eliminate rework and promote a zero-defect culture.
Completing such an improvement project would achieve the following organizational goals, key perfor- mance indicators, and deployment strategies:
• Significant cost savings • 20 percent waste reduction/five years • No negative audit findings • 10 percent cost-of-poor-quality reduction • 20 percent increase in schedule attainment • 75 percent decrease in backorders • Promote zero-defect culture
After careful evaluation, the project was formally selected in late 2011, and work began in January 2012.
Forming the Team
The company’s operational excellence training pro- gram helps develop high-potential Green and Black Belt candidates while also providing general aware- ness and skills training for other staff members. This program prepares employees to join various project improvement teams. A diverse group representing multiple disciplines was named to the improvement team, including:
Improvement team
Table 2 – Quality tools helped identify potential improvement opportunities in the cleaning process.
Tools used to identify possible improvements
What data was analyzed How analysis was performed
Process map Flowchart of entire process including cycle times
Visual observation to detect excessive complexity
Waste walk Facility and process were observed during operation
Team documented areas of waste, prompted by a standardized form
Brainstorming Group knowledge and experience Team and technical subject matter experts met to identify potential improvements
Benchmarking Industry standards and practices Subject matter experts provided insight into most current solutions
Root cause analysis
Process history and flowchart Causal relationships between inputs and outputs were identified
ASQ www.asq.org Page 3 of 5
improvements cited in the waste walk, five were significant enough to war- rant implementation. The team also utilized root cause analysis to determine which improvements might be most beneficial in solving the bottlenecks in the cleaning process.
Developing Strategies
To bring greater focus to possible improvement actions, three evaluation methods were used: value stream mapping (VSM) (see Figure 1), screening experiments, and benchmarking.
The maps clarified the steps needed to achieve the project’s objectives. They also verified the location of the bottlenecks in the process and where nonvalue-added waste was occurring, such as the time needed to clean out the tanks.
The value stream mapping for the current process allowed the team to determine that an entirely new procedure was necessary to satisfy the project’s objectives. Team members developed a five-part strategy, which consisted of the following changes:
• Converting from a batch process where each tank was filled and dumped sequentially to a continuous process where tanks could be jet-sprayed with a tank- cleaning machine (from Gamajet).
• Externalizing process tasks such as using a heat exchanger to provide hot water, instead of filling the tanks, then heating the entire volume; and draining the tanks continuously while cleaning, instead of having to wait until agitation was complete before draining.
• Separating sample collections. • Utilizing other water sources. • Standardizing the process, labeling the lines, and providing more detailed
batch records.
Sprayball
Screening experiments for differ- ent tank-cleaning machines plotted the cleaning efficacy against time to determine whether a specific
sprayball could achieve the required objectives. The data collected from these experi-
ments prompted the purchase of multiple sprayballs and manways (used to access the tanks). Finally,
benchmarking activities allowed team members to learn about past successes
to create a foundation for determining the best solutions to decrease the clean- out cycle time.
Gaps were analyzed between the current and future state based on obser- vations made during a changeover time analysis. Eleven tasks were identified, reducing the cleanout cycle time from 21 hours to 3.5 hours by implementing the sprayball technique to blast away contaminants and residues via high- impact cleaning jets.
The team believed if they could imple- ment a high-impact tank-cleaning machine with a continuous water stream, their solution would match the project’s objectives and perhaps provide even greater benefits. A new rapid, continuous cleaning procedure using a sprayball would address the fol- lowing items:
• The long, 64-hour changeover time • The low, 84 percent system
availability • A task ratio of 60 internalized to
zero externalized tasks • An operational equipment
effectiveness measure of 75 percent
Selecting Final Solutions
To gauge the effectiveness of the proposed improvements, team members collaborated with stake- holders to compare possible results of these actions against organiza- tional goals. They determined the
Figure 1 — Current state VSM
ASQ www.asq.org Page 4 of 5
improvements—particularly cycle-time reduction and improved first-time right percentage—would have a positive effect.
Quality tools played a major role in the selection of the final improvement actions, including:
• Failure mode effects analysis (FMEA) to analyze what issues could arise with the improvements and what corrective actions were necessary to mitigate any possible concerns.
• A cause-and-effect matrix narrowed the improvement list to a manageable number of options.
• Engineering studies verified what sprayball option provided the best, sustainable cleaning results.
The data analyzed included the specification needs of the tank- cleaning machine and what benefits it could produce under the projected conditions of use. The tank manway improvements would need to withstand a specific temperature and provide a perfect fit for the sprayball to clean the 2,000-gallon tanks effi- ciently. In addition, the batch record cycle-time data was studied for ways to reduce steps, or shift steps, so specific steps could be completed simultaneously. Finally, piping changes were needed in the manufacturing setting so specific dimensions were analyzed to ensure that the proposed changes would indeed work.
Reducing Cleaning Time, Saving Money
The five improvement actions—sprayball utilization, new man- ways, new batch records, standardized work, and externalized process tasks—were implemented as a week-long kaizen activ- ity. Duhigg, an ASQ Certified Six Sigma Black Belt (CSSBB) and Quality Auditor (CQA), says planning and the use of detailed to-do lists were the keys to success during the week of intense activity. During the kaizen week, the tank-cleaning machine was qualified and the procedure was optimized. All steps were standardized and the work was then captured in a formal batch record to ensure the operation would be performed in the same manner every time. This standardization would drive the improvement actions to be sustained over time.
The new procedure, combined with the standardized work, offers a means to ensure that the equipment is cleaned effec- tively each time. Since the implementation of the improvements, the average cleanout time (including two off-line tanks, which could not be upgraded) is 27 hours compared to the previous average of 64 hours, as shown in Figure 3.
Results
The reduced changeover time helped the Mallinckrodt focus factory to introduce an additional product, not originally pro- duced on this line. This added $700,000 in increased absorption, or new product manufactured, as shown in Table 3.
Table 3 – Results attributed to the improvement project were substantial.
Metric Baseline Results Change
Steam usage $938.70/year $469.35/year $469.35
Water/sewer usage $2,577.60/year $2,062.08/year $515.52
Additional production hours N/A +360 hours +$308,000/ 360 hours
Additional absorption N/A +$700,000 +$700,000
Total yearly savings N/A N/A $1 million+
260,000
255,000
250,000
245,000
240,000
235,000
230,000 Current rate/month with
2.7 day cleanout Future rate with cleanout time
reduction of 1.1 days
240,240
254,848
Line balance chart – Rate per month
Figure 2 — Cleanout time reduction
Summary: The takt time was being met by 240 lbs. at the current state production rate and current cleanout rate. By reducing the cleanout time with the implementation of the spray balls and standard work the building would be able to create 1.6 days of extra processing per month, which yield approximately 430 extra lbs./day or 13,000 lbs./month.
Figure 3 — Cleanout times (2010 to 2013)
160
140
120
100
80
60
40
20
0
H ou
rs
1/ 17
/2 01
0 2/
17 /2
01 0
3/ 17
/2 01
0 4/
17 /2
01 0
5/ 17
/2 01
0 6/
17 /2
01 0
7/ 17
/2 01
0 8/
17 /2
01 0
9/ 17
/2 01
0 10
/1 7/
20 10
11 /1
7/ 20
10 12
/1 7/
20 10
1/ 17
/2 01
1 2/
17 /2
01 1
3/ 17
/2 01
1 4/
17 /2
01 1
5/ 17
/2 01
1 6/
17 /2
01 1
7/ 17
/2 01
1 8/
17 /2
01 1
9/ 17
/2 01
1 10
/1 7/
20 11
11 /1
7/ 20
11 12
/1 7/
20 11
1/ 17
/2 01
2 2/
17 /2
01 2
3/ 17
/2 01
2 4/
17 /2
01 2
5/ 17
/2 01
2 6/
17 /2
01 2
7/ 17
/2 01
2 8/
17 /2
01 2
9/ 17
/2 01
2 10
/1 7/
20 12
11 /1
7/ 20
12 12
/1 7/
20 12
1/ 17
/2 01
3 2/
17 /2
01 3
3/ 17
/2 01
3
Cleanout times (2010 to 2013) FTR% since new cleanout: 100%
Cleanout time
New cleanout
introduced
ASQ www.asq.org Page 5 of 5
The results had a direct impact on organizational goals as the improvements reduced the number of cleaning failures nearly to zero, eliminating as many as two to three cleaning reworks per month. The project goal of promoting a zero-defect culture with a first-time right measure of 100 percent was reached, showing a 62 percent improvement. Also, nearly $2 million in backorders were completely eliminated at this focus factory, which led to the inclusion of a new product into the schedule. In addition to the tangible financial benefits, the newly installed manways not only improve the facility’s current good manufacturing practices, but also eliminate the potentially dangerous situation in the pre- vious process, when 2,000-gallon tanks full of water would be heated to boiling and agitated, then abruptly dumped.
Duhigg said the company is now planning on implementing improved cleaning processes throughout the site. “This project was essentially a pilot for our 20 different focus factories,” she said. “We are now going through and doing the same thing, across the entire 44-acre St. Louis facility.”
Earning Recognition
This project’s inclusion in the ASQ International Team Excellence Awards (ITEA) competition actually began on somewhat of a whim. As a new ASQ member, Duhigg received an email about the annual competition hosted at the World Conference on Quality and Improvement (WCQI), and was immediately intrigued. With support from Mallinckrodt’s Site
Leadership Team, and especially Operational Excellence and Quality management, the team assembled their application. The ASQ judges approved, and the project was named a finalist in the 2012 competition. In May 2013, Duhigg and her colleague Josh Blough presented the project at WCQI, along with repre- sentatives of 31 other teams from around the world. She said this project was a great match for the ITEA Process because of the value it delivered to the organization, “The benefits were so over-the-top, extravagantly obvious—$1 million a year,” Duhigg said, “the only question was, ‘Why didn’t we do it sooner?’”
For More Information
• To learn more about Mallinckrodt, visit www.mallinckrodt.com
• For more about the Gamajet spray machine, visit www.gamajet.com.
• For details on the International Team Excellence Award Process, visit wcqi.asq.org/team-award/.
• To read more examples of quality success, visit the ASQ Knowledge Center Case Studies landing page at asq.org/knowledge-center/case-studies.
About the Author
Janet Jacobsen is a freelance writer specializing in quality and compliance topics. A graduate of Drake University, she resides in Cedar Rapids, IA.