International Journal of Lean Six Sigma

Lean six sigma at a material recovery facility: a case study

Matthew Franchetti Pukhraj Barnala
Article information:
To cite this document:
Matthew Franchetti Pukhraj Barnala, (2013),"Lean six sigma at a material recovery facility: a case study",
International Journal of Lean Six Sigma, Vol. 4 Iss 3 pp. 251 - 264
Permanent link to this document:
http://dx.doi.org/10.1108/IJLSS-05-2013-0026
Downloaded on: 05 May 2015, At: 03:00 (PT)
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

References: this document contains references to 17 other documents.

To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 258 times since 2013*
Users who downloaded this article also downloaded:
Bernardo Nicoletti, (2013),"Lean Six Sigma and digitize procurement", International Journal of Lean Six
Sigma, Vol. 4 Iss 2 pp. 184-203 http://dx.doi.org/10.1108/20401461311319356
Nurul Fadly Habidin, Sha'ri Mohd Yusof, (2013),"Critical success factors of Lean Six Sigma for the
Malaysian automotive industry", International Journal of Lean Six Sigma, Vol. 4 Iss 1 pp. 60-82 http://
dx.doi.org/10.1108/20401461311310526
Lucila M.S. Campos, (2013),"Lean manufacturing and Six Sigma based on Brazilian model PNQ: An
integrated management tool", International Journal of Lean Six Sigma, Vol. 4 Iss 4 pp. 355-369 http://
dx.doi.org/10.1108/IJLSS-08-2012-0007

Access to this document was granted through an Emerald subscription provided by 235366 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.

*Related content and download information correct at time of download.

 The current issue and full text archive of this journal is available at
www.emeraldinsight.com/2040-4166.htm

Lean six sigma

Lean six sigma at a material at MRF
recovery facility: a case study
Matthew Franchetti and Pukhraj Barnala
Mechanical, Industrial and Manufacturing Engineering Department, 251
The University of Toledo, Toledo, Ohio, USA

Abstract
Purpose The purpose of this paper is to discuss the lean six sigma implementation process and its
quantified benefits for the recycling industry. Specifically, the goals of the project were to improve the
processes and increase capacity for a government operated material recovery facility (MRF) in Toledo,
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

Ohio, USA.
Design/methodology/approach The six sigma DMAIC methodology was utilized for this study.
This included aligning and optimizing processes and the removal of process-generated defects and errors.
Findings After comparison of the current and future states of the MRF, it is found that by reducing
non-value added activities, productivity was increased by 7.3 percent for paper bales, 12.8 percent for
commingled bales and 1.6 percent for old corrugated container (OCC) bales. The project generated over
$65,000 in annual savings for the facility.
Research limitations/implications The findings are limited due to the focused nature of the
case study and further cost-benefit analysis can be carried out.
Practical implications This paper may be useful to researchers and practitioners for
understanding lean six sigma implementations and its derived benefits for the recycling industry
and government operated processing facilities.
Originality/value The paper is a real case study showing lean six sigma implementation and its
benefits for the recycling industry; a field that has not been explored to a high degree.
Keywords United States of America, Six sigma, Recycling, Process improvement,
Material recovery facility, Materials management, DMAIC cycle, Waste recovery
Paper type Case study

1. Introduction
The purpose of this recycling research was to study the application of lean Six Sigma
principles in government controlled operations for municipal recycling collection and
processing. Lean Six Sigma has not been used in the solid waste reduction or
environmental protection fields to the extent that it has been applied in manufacturing
and in other sectors (Banuelas et al., 2005). Lean Six Sigma has been applied to a variety
of industries including pharmaceutical (Goodman, 2012), assembly automation
(Lee-Mortimer, 2006), developing countries (Malik and Blumenfeld, 2012), and
telecommunications (Psychogios et al., 2012). A review of recent literature indicated
increased research activities in the application of lean Six Sigma to non-manufacturing
fields such as health care operations (Van den Heuvel et al., 2004) and to call centers
(Laureani et al., 2010), but few detailed case studies related recycling. The objective of
this research was to investigate the application and benefits of lean Six Sigma for this International Journal of Lean Six
industry. Specifically, the objective of the project was to improve the processes and Sigma
Vol. 4 No. 3, 2013
operations for a material recovery facility (MRF) located in Toledo, Ohio, USA that was pp. 251-264
operated by the local government. This included aligning and optimizing processes and q Emerald Group Publishing Limited
2040-4166
removal of process generated defects and errors. These improvements primarily focused DOI 10.1108/IJLSS-05-2013-0026
 IJLSS on the reduction of municipal solid waste (MSW) entering local landfills; optimization
4,3 cycle times for the MRF; enhancement end-user/customer satisfaction; efficiency
improvements; cost reduction; and the elimination of errors/defects. The intent of these
objectives were two fold; first to demonstrate that lean Six Sigma techniques can be
successful applied in the material recovery industry and second that these techniques
can be successfully applied in a government controlled operations. Sub-objectives
252 included developing process standards to eliminate bottlenecks and redundancies. Lean
Six Sigma was selected for this study to address both cost and quality issues that the
facility was experiencing. Due to lower tax revenues, the government operated facility
was faced with cost reductions to meet budget targets at the time of the study.
Additionally, recycling mills that and processors that received the end product from the
facility were dissatisfied with the level of contamination in paper and plastic bales. This
project was chosen to address these issues by lowering operating costs and enhancing
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

customer satisfaction by reduction defects. At the time of this study, the MRF
administration team was preparing to add additional equipment to process additional
recyclable materials collected from nearby cities and towns. The MRF administration
team established a goal to become more productive and efficient to better utilize tax
dollars and serve the public. From the MRF administration point of view, this research
aided in enhancing the understanding of the current operations, established a sigma
level for the processes, identified improvement opportunities, identified cost reduction,
and solutions to reduce defects per million opportunities (DPMO). From a research and
originality standpoint, this study provides a resource for those involved in the areas of
recycling and government as related to lean Six Sigma; an area that has not been as fully
studied as manufacturing operations.
This case study follows the five step methodology used in the Six Sigma process,
the DMAIC cycle. DMAIC stands for define, measure, analyze, improve, and control
(Pande et al., 2000). The define step is outlined in Section 2. In Section 2 the problem is
identified and specific goals are determined. The measure step focuses on gathering
raw data from the process. This is described in detail in Section 3.1. The third step,
analyzing the data will be shown in Section 3. This is an analysis of the gathered data
and how it is meaningful for the MRF. In Section 4, the suggestions for the company
are explained in more detail. Several improvement opportunities give possible methods
to improve the printing processes and fulfill the fourth step in the Six Sigma process.
Finally, the methods for sustaining the changes, (Step 5) are discussed.

1.1 Project goal

The goal of this case study was to reduce process waste, reduce defects, and increase
production capacity to meet projected increased capacity needs for the Lucas County
MRF. Six Sigma and lean practices were applied to ensure that the changes will be
effective and sustainable. To begin the study, a project charter was created that
identified the project team, goals stated above, start date, and expected end date.

2. About the case study

A MRF is an operation where recyclable materials are received, sorted, and prepared
for sale to end-users or buyers to create new raw materials. There are two types of
MRFs, clean and dirty. A dirty MRF processes raw MSW to extract and sort recyclable
materials. A clean only processes and sort commingled recyclable materials with
 minimal MSW contamination. Figure 1 shows a flowchart of a typical clean MRF Lean six sigma
operation. The MRF operation analyzed for this study involved a facility employs at MRF
30 and is owned and operated by the local government. The lean Six Sigma team
consisted of the facility manager, supervisor, baler operator, maintenance employee,
and a team of three graduate students from the Industrial Engineering Department at
the University of Toledo. This project represented the first initiative related to lean
Six Sigma for the facility. 253
As shown in Figure 1, a clean MRF processes commingled recyclable generated by
local residents and businesses. MRFs are designed such that the input materials or
recyclables from curb side recycling programs can be separated and processed into
marketable commodities. The MRF discussed for this case study is a basic multi-stream
MRF with manual sorting operations. As shown in Figure 1, a multi-stream MRF
separates and bales commingled containers including ferrous metals, aluminum cans,
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

polyethylene terephthalate (PET) and high-density polyethylene (HDPE) bottles, mixed

glass old corrugated containers (OCC), mixed office paper (MOP), old newspaper (ONP)
and magazines. Non-recyclable and residue from both the streams after manual sorting
are collected and sent to landfill. After the manual sortation at sort station A and B as
shown in Figure 1, these materials are bailed using a baler and finally ready for shipment
to the end-users.
This case study examines a MRF located in Toledo, Ohio, USA. The MRF is owned
by a division of the local government, the Lucas County Solid Waste Management
District (LCSWMD). At the time of the study, the MRF was used to sort and sell
approximately 10,000 tons of recyclable material annually. The LCSWMD funds and
maintains approximately 174 drop off recycling sites throughout Lucas County. The
drop-off sites are located at apartment complexes, schools, large grocery stores, and
metro parks. Each drop-off container is owned and maintained by the LCSWMD. Daily
trucks are dedicated to pick commingled fiber and container stream. As the truck
reaches its capacity they unload the material at the MRF for sortation and bailing.
Before unloading materials trucks are weighed and recorded for payment plus to keep
a track of the system. A ticket is generated for each truck and total weight is recorded.
The facility has a sortation center, a baler and a warehouse for shipping final products
to end-users. Plate 1 shows a picture of the MRF sortation process.

Figure 1.
Flowchart for a basic
clean MRF operation
 IJLSS
4,3

254

Plate 1.
Sorting operation for the
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

Lucas County MRF

In terms of process efficiency, the MRF operations were not reaching its full potential.
Large variations in daily processing rates were observed in conjunction with large
inventories of unprocessed materials. The principles of lean and Six Sigma are useful
to the manufacturing process in order to maximize the production output while
reducing the costs associated with defective parts (Henderson and Evans, 2000).
Additional benefits include increased utilization of equipment and labor, improved
safety, maximization of floor space utilization, and reduced inventory. These benefits
will allow the MRF to increase efficiencies by allowing them to reach their projected
production volume to best utilize tax dollars. In addition, efficiency enhancements and
variability reduction will allow the MFR to increase its capacity with minimal
additional capital expenditure. This was a critical need for the MRF, since plans were
being developed to begin the processing of an additional 5,000 annual tons of
recyclable materials from a local communitys curbside collection program. The
following sections will discuss the baseline measures, including the DPMO, analysis,
improvements, and control strategies.

3. Application of the Six Sigma define, measure, analyze, improve and

control methodology
3.1 Define
The process overview provides a basic understanding of the entire system and offered a
preliminary identification for defects and variation in the process. The process begins
from curbside pickup of commingled containers including HDPE and PET plastics,
glass containers, aluminum can, MOP, OCC, newspaper and magazines. Once the
recyclable materials from the LCSWMD owned dumpsters are collected, the district
truck transports the materials to the MRF which serves as the raw material for the
process. As the trucks enter the MRF to unload, a ticket is generated which displays the
weight of the truck. The ticket provides the amount of recyclable material that was on
the truck. After that materials are separated from each other by manual sortation of
material from three to four employees. Each material is separated and stored in a box or
on a pallet. Once these pallets are full, the pallets are moved near the baler area, so as to
start the bailing operation after sortation. The recyclable materials are then bailed and
stored in a warehouse for shipping them to the final destinations or customers.
 3.2 Measure Lean six sigma
The purpose of the measurement phase was to baseline the existing processes to at MRF
establish current-state metrics (Breyfogle et al., 2000). Several weeks of on-site data
collection at the MRF were required. Some of the tools used in the measure phase of the
project were Pareto analyses, time studies, surveys, interviews, and check sheets.
3.2.1 Pareto analysis. A Pareto chart is a vertical bar graph that displays problems or
improvement opportunities in a prioritized order, so it can be determined which 255
problems or opportunities should be addressed first. It is a tool used on the 80:20 rule
which means usually 80 percent of the problems come from 20 percent of the processes.
The categories are sorted in decreasing order from left to right in a Pareto chart by their
count or cost, whichever is displayed (Pyzdek and Keller, 2003). Figure 2 shows the
Pareto analysis for the amounts (in tons) of materials sorted per year at the MRF.
The annual recyclable material to be processed through MRF was 11,432 tons. From
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

the Pareto chart, the most collected material at the MRF was ONP, MOP and then OCC
and aluminum cans.
3.2.2 Time studies. Time studies are the continuous surveillance of a job/process, using
a timekeeping device (stopwatch used in this project) to record the time taken for
accomplishing a job. After recording the time, the workers performance time (level) is
recorded and then data is used to make the standard time for the task/process. Standard
times (minutes/cycle) for the baler was calculated from data observations for each material.
A sample of the data collected for the cycle time to bale a load of MOP is displayed in Table I.
Table I displays the average standard time taken for a bale of paper to process
through the baler. The average time to process and bale the MOP was 6.2 minutes with a
standard deviation of 0.95. The average time to process a bale of OCC was 6.7 minutes
with a standard deviation of 1.10. The variation in the system was due to breakdowns of
cable system that cover the bale, removal of unwanted glass particles coming out from
baler, and other continuous maintenance operations. Time studies were also conducted
to determine transportation cycle times to transport the bales from the baler to the
warehouse staging area. The average cycle time to transport the bales was observed to
be 1.8 minutes.

Figure 2.
Pareto analysis
 IJLSS
Time taken for a bale of paper to process
4,3 through the baler
Time in Time out Total time for process (mins)

0.0000 0.0812 4.9

0.0812 0.2787 11.9
256 0.2787 0.3320 3.2
0.3320 0.3799 2.9
0.5400 0.6600 7.2
0.6070 0.6689 3.7
0.6835 0.7924 6.5
Table I. 0.3799 0.5400 9.6
Time study data Average std. time 6.2
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

3.3 Analysis
To begin the analysis phase, a fish bone diagram was created to better understand
sources of variation and delay in the processes as shown in Figure 3. As shown in
Figure 3, material, machinery, methods and staff were the main factors for variation and
delay. From a methods standpoint, employee surveys and interviews were used to
collect data. These surveys revealed that the manual sorting method used could be
improved to reduce the overall defects. Other comments included issues regarding
complexity of the process, limited quality control, and lack of standard operating
procedures.
The primary issue with the staff was related to unscheduled work absences, if a
scheduled operator was not present it became very difficult for another operator to
cover the shift, due to lack of standards and contingency plans. The staff did appear to
require additional training regarding the defects in the process.
As shown in Figure 3, raw material caused significant variability in the system. The
main reason was related to broken glasses as it can contaminate the end products. OCC
is also one of the major defects in the raw material stream based on condition and
quality. Shredded paper is also one of the main defects because it is difficult to bale.

Figure 3.
Fishbone diagram
 Another source of variation in the process was related to machine breakdown and Lean six sigma
improper resource utilization. For example, a single baler was used for all materials at MRF
types. Due to improper maintenance and extensive use of baler, the cable system that
wraps the bale tended to break and malfunction frequently. There was a large amount
of non-value added time lost in fixing the cable wiring system.
3.3.1 Machine utilization analysis. To better understand the equipment usage, the
equipment requirements were determined via a three on-site data collection audit. 257
The amount of equipment required for a process is referred to as the equipment fraction.
The equipment fraction can be determined for a process by dividing the total time required
to perform the process by the time available to complete the process. The total time
required to complete the process is product of the standard time for the process and the
number of times the operation/process is to be performed (Tompkins and White, 1984).
Standard time was then calculated by using time studies and other statistical tools:
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

 
1
F S * Q*
H *E *R

where:
F number of machines required per shift to perform the task.
S standard time (minutes) per unit produced.
Q number of units to be produced per shift.
H amount of time available per machine.
E actual performance, expressed as a percentage of standard time.
R reliability of machine, expressed as percent up time.
Additionally, equipment requirements are a function of the factors such number of shifts
per day, set up time, degree of flexibility, layout type and maintenance. Standard cycle
times were measured for current baler per material to determine resource utilization. As
shown in Table II, the baler was used to process 12,000 bales per year. The baler was
estimated to be 85 percent reliable based on the machine utilization audit. The efficiency
in terms of standard cycle time was measured to be 90 percent which means that

Average SD Bales
cycle of Annual Pounds per Available Machine Machine
time cycle tons to be per year time per reliability Efficiency fraction
Material (min.) time processed bale (Q) year (H) (R) (E) (F)

ONP 6.2 0.95 3,759 1,630 4,612 120,000 85.00% 90.00% 0.31
OCC 6.7 1.10 3,410 1,700 4,012 120,000 85.00% 90.00% 0.29
MOP 6.2 0.80 1,356 1,630 1,664 120,000 85.00% 90.00% 0.11
Plastic 7.1 1.10 793 1,200 1,322 120,000 85.00% 90.00% 0.1
Steel can 7.1 0.99 275 1,830 301 120,000 85.00% 90.00% 0.02
Aluminum Table II.
can 7.1 0.95 82 1,830 90 120,000 85.00% 90.00% 0.01 Equipment requirement
Total 9,675 9,820 12,000 0.85 by material for the baler
 IJLSS machine is processing at 10 percent less than its standard. Based on standard cycle times
4,3 and 250 working days of 8 hours shift a day, the baler was utilized 85 percent.
3.3.2 DPMO. DPMO are a measure of process performance or non-conformities
per million opportunities. A defect is also known as a non-conformance of quality
characteristic to its requirement and specification. For this study, defects were defined
as tons of material by category that were received at the MRF that could not be
258 recycled due to improper condition or contamination. Defects for the processes
are shown in Table III with the annual weights of defects. As displayed in Table III, the
main defects of the process are glass particles which amounted to 85 tons per year.
Based on this analysis, DPMO for the process was calculated as 2,788 DPMO which
correspond to a sigma level of 4.3 in terms of weight.
3.3.3 Cost analysis. The annual utility/maintenance expenses for the MRF were
calculated and estimated to be $95,000 and annual labor expenses of $820,000.
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

A revenue analysis and an average dollar amount per ton for each recyclable material
were also calculated and are displayed in Table IV. The total annual revenue generated
from the sale of recyclables based on the average commodity rates was $1,012,807.
This cost analysis served as an initial baseline for process improvements based on lean
Six Sigma concepts.

3.4 Improve
The next phase of the study involved improving the system to reduce variability and cost
by removing major bottlenecks and sources of defects (Park, 2002). Value added and

Component Tons per year Percent of total Tons recycled Tons not recycled

Glass 31.85 94.1 0.00 31.85

Mixed office paper 1.37 4.1 0.00 1.37
Non-recyclable/food 0.22 0.7 0.00 0.22
Cardboard 0.15 0.4 0.00 0.15
HDPE (2) 0.13 0.4 0.00 0.13
Aluminum cans 0.03 0.1 0.00 0.03
Newspaper 0.00 0.0 0.00 0.00
PET (1) 0.10 0.3 0.00 0.10
Magazines 0.00 0.0 0.00 0.00
Table III. Plastic wrap 0.00 0.0 0.00 0.00
Annual weight of Other 0.00 0.0 0.00 0.00
solid waste or defects Other plastics 0.00 0.0 0.00 0.00
from the MRF Total 33.85 100.0 0.00 33.85

Material Dollar per baled ton Annual tons to be processed Annual revenue

ONP 85 3,759 319,515

OCC 110 3,410 375,100
MOP 82 1,356 111,192
Plastic 180 793 142,740
Table IV. Steel can 180 275 49,500
Total revenue generated Aluminum can 180 82 14,760
from recycling Total 9,675 1,012,807
 non-value added analyses were a major part of this phase. Table V displays the dollar Lean six sigma
loss per year due to residue in the system that cannot be sold on the commodity market, at MRF
which amounted to nearly 35 tons of material and $1,000 per year. Upon reviewing the
data, major improvement areas were broken glass particles, contaminated OCC, baler
machine failures, time wasted in transportation of baled material to warehouse and loose
shredded paper.
3.4.1 Value added vs non value added analysis. Lean manufacturing is a systematic 259
approach to identifying and eliminating waste which is also known as non-value added
activities through proper continuous improvement techniques (Pande and Holpp,
2002). The three factors that were evaluated for work to be considered as value-added
were (Yang, 2005):
(1) Capacity machinery, resources, tools and employees used in the process must
have the necessary capacity to produce a final product adding value to for this
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

study, utilization for baler was considered.

(2) Information/instructions employees must know their final product and
process to achieve the final product with minimum waste or non-value added
activities. For the MRF, all workers should have a thorough understanding of
the overall process plus they should very well understand the difference
between recyclables and non-recyclables.
(3) Materials the material given to workers should be free from defects and able
to be processed and sold on the commodity market. A worker should know
which raw material is acceptable or non-acceptable.

The five sources of waste that were identified and analyzed for this study, were
adapted from previous literature (Kwak and Anbari, 2006) and included:
(1) Defects defects are due to various reasons starting from material, inadequate
training, labor, sorting or deficient maintenance. For the MRF operation, defects
were mainly due to baler breakdowns, defects in raw materials, and incomplete
engineering specifications. At the MRF, the time studies indicated that
29 percent of the non-value added time is due to repairs for the cable wire
system which wraps the final baled products.
(2) Transportation unwanted movement of materials within the facility makes it
a waste or non-value added activity from the customers viewpoint. Several
examples observed at the facility included multiple movements of MOP or OCC
to several staging areas near the baler. A single movement from the unloading
zone to the baler would reduce this form of waste.

Material Tons per year Dollar per baled ton Annual dollars lost ($)

Glass 31.85 25 796.25

Mixed office paper 1.37 82 112.34
Cardboard 0.15 110 16.5
HDPE (2) 0.13 180 23.4
Aluminum cans 0.03 180 5.4
Newspaper 0.00 85 0 Table V.
PET (1) 0.1 180 18 Dollar loss per year
Totals 33.63 971.89 due to defects
 IJLSS (3) Waiting the time wasted such as waiting for raw material, waiting time for
equipment or tool set ups and instructions, waiting for tooling, and waiting for
4,3 workers. This all adds up to a large amount of non-value added activities to the final
product. The waiting total time observed during the time studies was 1.1 minutes
per 8.1 minutes cycle (13.6 percent for the total cycle time). Wait time issues included
delays related for next process to start, idle time between two processes, and time
260 wasted during miscellaneous activities including restrooms breaks. Wait time
represented 47 percent of the non-valued added activities per cycle.
(4) People time wasted in not utilizing the abilities of a worker or employee to
their fullest potential describes this form of waste. Morale and company culture
is also one of the main factors for this kind of waste. One example of this form of
waste observed at the facility included work activities of the trained baler
operator. Several instances were observed that involved him doing lower level
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

work such as sweeping the floors or operating a fork lift.

(5) Motion extraneous motion are poor plant layout and workplace design
contributed to motion losses. It included time wasted in searching for tools,
extra product handling, or material stacking. For the MRF, time taken by a
worker to move baled items from the baler to the shipping area were included in
this type of waste.
The main waste or non-value added activities in our case study were the time required
to fix the baler cable system, which occurred seven to eight times a day.
3.4.2 Capacity. Due to changing demands of a product an organization needs to
determine the production capacity. For the MRF, the decision to process additional
recyclable material from nearby communities related directly to capacity. Capacity can
be increased by using modern manufacturing methods, continuous improvement in
facility, optimum utilization of machines and workers and there are so many other
process improvement techniques depending upon the business to increase efficiency
and reduce waste. The formula used for calculating the tons per hour of a material
processed at the MRF was:
 
1
Tx Bx*Wx*
2000
where:
Tx tons per hour processed for material x.
Bx number of bales processed in one hour for material x:
 
1
Bx Nx*60* :
Px
Nx number of bales processed for material x in total time.
Px total time taken to process the material.
Wx pounds per bale for material x.
Table VI displays the capacity of the system under its current state in terms of
processing capability per hour per material type.
 Tons per hour processed for MOP, containers, and OCC were 7.5, 6.5 and 7.6 tons Lean six sigma
per hour, respectively. The capacity of processing each material per hour can be at MRF
increased be reducing the non-value added time and waste from the process. By
implementing the recommendations that were identified, the tons per hour processed at
the MRF could reach a higher level. Table VII represents the tons per hour that could be
processed assuming the recommendations were implemented. After implementing
the recommendations there will be at least an increase in tons processed per hour 261
by 7.3 percent in paper bales, 12.8 percent in commingled bales and 1.6 percent in
OCC bales.

3.5 Control
The purpose of the control phase is to monitor and maintain improvements. Control
charts, documentation, and communication are the critical components of this phase.
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

To verify that improvements were achieved, a statistical analysis between the previous
and improved system states was analyzed. To be statistically confident that the mean
cycle time after the recommendations were implemented were lower than the baseline
means cycle times, a hypothesis test was performed. A t-test was conducted with the
following hypothesis for the mean cycle time of plastic bales:
H0 (null hypothesis).
m # 6.5, new mean cycle time is less than or equal to 6.5.
H1 (alternate hypothesis).
m . 6.5, new mean cycle time is greater than 6.5.
The sample size was 20 cycles for the each scenario (before and after
recommendations). The normality assumption for the t-test was validated using a
Kolmogorov-Smirnov test for each sample set. The F-test was used to validate the
equal variance assumption between the two populations. The samples were assumed
independent as they were taken before and after process changes. The test statistic for
this test was calculated to be 0.559. At a 5 percent significance level, the critical t-value
(ta) is 2.776.The t-test indicates that we do not reject H0. In other words it indicates

Number of Total time for processing the Pounds per Tons per hour
bales (Nx) Material (x) material (Px)-(mins) bale (Wx) processed (Tx)

9 Paper 58.6 1,630 7.5 Table VI.

5 Commingled 35.2 1,515 6.5 Tons per hour processed
9 OCC 61 1,700 7.6 for each material

Number of Total time for processing the Pounds per Tons per hour
bales (Nx) Material (x) material (Px)-(mins) bale (Wx) processed (Tx)

9 Paper 54.6 1,630 8.1 Table VII.

5 Commingled 31.2 1,515 7.3 Tons processed per hour
9 OCC 60 1,700 7.7 after recommendations
 IJLSS there is a 95 percent confidence level that the new mean is less than 6.5 minutes after
4,3 implementing the recommendations. Statistically, it can be concluded that new mean
cycle times are lower than the previous state cycle times.
3.5.1 Control charts. To maintain and monitor the improved cycle times, control charts
were created to identify special causes of variation, delays, or bottlenecks in the system
(McAdam and Lafferty, 2004). The control charts were created using a standard three
262 sigma control limit. A sample control chart for the cycle time to bale a load of OCC is shown
in Figure 4. The control charts utilized to monitor performance were the x and R charts.
The x and R chart were chosen since a process variable (cycle time) was being monitored
for small and constant samples of five. The samples were collected every hour.
The control charts were automatically generated per cycle once the operator pressed
the start button to begin a bale cycle. The control chart offered a real time
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

measurement for the leadership team at the MRF to monitor production rates and
identify delays quickly. For example, in Figure 4, the final sample is out of control due
to a baler wire jam. Special causes of variation were addressed by a team consisting of
the operator, maintenance team member, and the shift supervisor.
3.5.2 Process capability. The capability of the improved process was calculated by
measuring the tons processed per hour for each material. The final capability is
displayed in Table VIII and it indicated the before and improved levels. Costs related to
poor quality were also a significant. In terms of lost revenue, the costs of poor quality
were approximately $1,000 per year as displayed previously in Table V. The
preventative costs to remove these defects (broken glass mixed in MOP of OCC) were
minimal and addressed by proper handling and separation of glass materials.

Figure 4.
Control chart showing
time taken to process
the OCC bale

Before Improved levels

Tons processed per hour for paper 7.5 8.1

Tons processed per hour for OCC 7.6 7.7
Table VIII. Tons processed per hour for commingled 6.5 7.3
Final capability chart Total savings (annual) $65,019
 3.5.3 Managerial implications. The managerial implications from this study strongly Lean six sigma
indicate that lean process redesign had a significant impact on reducing operational at MRF
costs (approximately $65,000 per year). Maintaining the improved standard cycle to
process bales of material is critical to maintain these savings and the developed controls
will allow the managerial team to accomplish this task. Also, the cost of poor quality
related to broken glass could generate an additional $1,000 per year if proper separation
and handling and enforced. 263
4. Conclusions
This case study demonstrated that lean Six Sigma can be successfully applied in the
recycling and material recovery fields. In addition, it also demonstrated that these
concepts can be applied with success in government controlled operations. The case
study illustrated how the DMAIC approach of Six Sigma aided in identifying
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

bottlenecks and barriers to the operations of MRF. DPMO and sigma level calculations
for the current process were calculated to identify potential economic and environmental
benefits. By reducing non-value added activities, productivity was increased by
7.3 percent for paper bales, 12.8 percent for commingled bales and 1.6 percent for OCC
bales. The case study also enhanced employee participation in process improvement, the
use of statistical tools to solve problems, increased process knowledge, and it motivated
the MRF leadership team towards Six Sigma approach.
Key lessons learned from this study relate to identifying, implementing, and
maintaining efficiency improvements. By improving efficiency by 7-12 percent, the
$65,000 annual cost savings could be achieved. Consistent and accurate data collection
(process observations and time studies) over several days were necessary. Other
lessons learned related to the need to involve all team members (operators, managers,
and maintenance employees) in the decision and improvement process to foster buy-in.
From this data, improvement opportunities were analyzed and action plans
were created to help implement those plans. Standardized methods were developed
to ensure that the process would be sustainable. If the MRF leadership team is able
to successfully carry out the proposed suggestions, they will be able to increase
capacity and process additional recyclables from nearby communities. This will better
utilize tax dollars and prevent the need to purchase capital equipment to increase
capacity.
In terms of the DMAIC methodology, several modifications and specific tools were
utilized for this novel application related to recycling and MRF operations. Specifically,
a machine fraction requirement for the baler was determined before and after
recommendations. This machine fraction calculation offered insights regarding the
utilization of the single baler unit at the facility and aided the team in determining if
one baler was adequate to process current and future workloads. Additionally,
analyzing a single baler that processes multiple materials (OCC, MOP and mixed
containers for this case study) presented challenges in terms of allocating machine
usage, but was alleviated by use of the machine fraction calculation.
Limitations from this study relate to the geographical location and operational
characteristics of the facility. Although the results could apply to similar facilitates, the
case study used commodity market data from Chicago, Illinois and the facility size was
on the small to medium range (annual processing of 10,000 tons of material). Larger
facilities or facilities with more automation may yield different results.
 IJLSS References
4,3 Banuelas, R., Antony, J. and Brace, M. (2005), An application of Six Sigma to reduce waste,
Quality & Reliability Engineering International, Vol. 21, pp. 553-570.
Breyfogle, F.W., Cupello, J.M. and Meadows, B. (2000), Managing Six Sigma: A Practical Guide to
Understanding, Assessing, and Implementing the Strategy That Yields Bottom-Line
Success, Wiley, New York, NY.
264 Goodman, E. (2012), Information analysis: a lean and Six Sigma case study, Business
Information Review, Vol. 29 No. 2, pp. 105-110.
Henderson, K.M. and Evans, J.R. (2000), Successful implementation of Six Sigma:
benchmarking: general electric company, Benchmarking: An International Journal,
Vol. 7 No. 4, pp. 260-281.
Kwak, Y.H. and Anbari, F.T. (2006), Benefits, Obstacles, and Future of Six Sigma Approach,
Elsevier Science, New York, NY.
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)

Laureani, A., Antony, J. and Douglas, A. (2008), Lean Six Sigma in a call centre: a case study,
International Journal of Productivity and Performance Management, Vol. 59 No. 8,
pp. 757-768.
Lee-Mortimer, A. (2006), Six Sigma: a vital improvement approach when applied to the right
problems, in the right environment, Assembly Automation, Vol. 26 No. 1, pp. 10-17.
McAdam, R. and Lafferty, B. (2004), A multilevel case study critique of six sigma: statistical
control or strategic change?, International Journal of Operations & Production
Management, Vol. 24 No. 5, pp. 530-549.
Malik, A. and Blumenfeld, S. (2012), Six Sigma, quality management systems and the
development of organisational learning capability: evidence from four business process
outsourcing organisations in India, International Journal of Quality & Reliability
Management, Vol. 29 No. 1, pp. 71-91.
Pande, P. and Holpp, L. (2002), What is Six Sigma, McGraw-Hill, New York, NY.
Pande, P., Neuman, R. and Cavanagh, R. (2000), The Six Sigma Way: How GE, Motorola, and
Other Top Companies are Honing Their Performance, McGraw-Hill, New York, NY.
Park, S.H. (2002), Six Sigma for productivity improvement: Korean business corporations,
Productivity Journal, Vol. 43 No. 2, pp. 173-183.
Psychogios, A.G., Atanasovski, J. and Tsironis, L.K. (2012), Lean Six Sigma in a service context:
a multi-factor application approach in the telecommunications industry, International
Journal of Quality & Reliability Management, Vol. 29 No. 1, pp. 122-139.
Pyzdek, T. and Keller, P. (2003), The Six Sigma Handbook, McGraw-Hill, New York, NY.
Tompkins, J. and White, J. (1984), Facilities Planning, Wiley, New York, NY.
Van den Heuvel, J., Does, J.M. and Vermaat, M.B. (2004), Six Sigma in a Dutch hospital: does it
work in the nursing department?, Quality & Reliability Engineering International, Vol. 20
No. 5, pp. 419-426.
Yang, K. (2005), Design for Six Sigma and value creation, International Journal of Six Sigma
and Competitive Advantage, Vol. 1 No. 4, pp. 355-368.

Corresponding author
Matthew Franchetti can be contacted at: matthew.franchetti@utoledo.edu

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

Or visit our web site for further details: www.emeraldinsight.com/reprints
 This article has been cited by:

1. Marcus Assarlind, Lise Aaboen. 2014. Forces affecting one Lean Six Sigma adoption process. International
Journal of Lean Six Sigma 5:3, 324-340. [Abstract] [Full Text] [PDF]
Downloaded by Western Michigan University At 03:00 05 May 2015 (PT)