The article describes a project that employs problem-based learning (PBL) to teach the Lean Six Sigma (LSS) methodology as part of an undergraduate or graduate business course. It is scalable to a variety of course delivery and schedule formats, and uses data sets that can create distinct problem-solving scenarios for up to 16 student teams. It exploits available internet resources for supplemental information, thus requiring no physical setup or experimentation, and eliminating material costs. The project provides a complex business problem in a familiar setting, allowing students to focus their learning efforts on technical mastery of concepts and tools rather than underlying business descriptions. It emphasizes critical thinking, team work, and project management, in addition to the fundamental concepts and supporting tools of the process improvement method.
Subject Areas: Problem Based Learning Pedagogy, Lean Six Sigma, Supply Chain Management Pedagogy, and Process Improvement Case Study.
INTRODUCTION AND CONCEPTUAL FOUNDATIONS
This article describes a project that employs problem-based learning (PBL) to teach the Lean Six Sigma (LSS) methodology as part of an undergraduate or graduate business course. The project is executed as a multiphase, course-long assignment to approximate the just-in-time learning format that is common in LSS training programs (Breyfogle III, 2008) in which learning material is delivered in stages that roughly correspond to a project phase, and time in-between learning segments is spent applying that learning to a chosen project.
The LSS methodology, commonly referred to as the DMAIC (Define, Mea- sure, Analyze, Improve, and Control) approach, is a multiphase methodology with
Miller, Hill, and Miller 383
key objectives for each phase. Lean aspects of the methodology focus on iden- tifying nonvalue added activities (waste), and driving efficiency by eliminating pure waste and externalizing (or minimizing) required waste. Six Sigma aspects focus on identifying process elements that generate process variation and result in defects, and increasing efficiency by reducing or eliminating sources of variation (Pyzdek & Keller, 2014). Goldsby and Martinchenko (2005) described the LSS methodology as an enabler for supply-chain processes.
PBL is an open-inquiry method in which students work in small groups on problems or scenarios that are typically large, complex, and ill-defined. These con- ditions are precisely what LSS practitioners face when trying to resolve a business problem. A PBL problem requires students to work together to first identify what is needed to solve it, engage in self-directed learning to obtain needed information, use acquired knowledge to solve the problem at hand, and then reflect on what has been learned and learning strategies. Instructor(s) and/or peer tutors facilitate the work of the groups by providing direction but not answers, scaffolding the students’ learning by coaching and modeling (Hmelo-Silver, 2004). PBL’s goals include the development of flexible knowledge, effective problem-solving skills, self-directed learning skills, effective collaboration skills, and critical thinking skills including learning how to find, evaluate, and use appropriate learning and informational sources, learning to work collaboratively and cooperatively, and de- veloping intrinsic motivation (Duch, Groh, & Allen, 2001; Hmelo-Silver, 2004). A benefit of bringing LSS into the classroom in the context of a multiphase, PBL project is that students receive practice with quantitative, qualitative, and logical analysis tools that reinforce critical thinking, introduce project management, and provide them with immediately relevant skills.
Numerous examples exist of using the LSS methodology in academic courses and degree-granting programs, and in diverse fields. For example, LSS concepts or applications in the sciences and engineering can be found in such areas as an undergraduate Nanotechnology course (Genis et al., 2010) and a Structured Prob- lem Solving class for master’s level students (Anderson-Cook, Patterson, & Hoerl, 2005). Within business schools, several courses teach LSS concepts while student teams work to solve real company business problems. Shafer (2005) described a program that integrates the American Society for Quality Six Sigma Black Belt certification preparation into an MBA curriculum and incorporated projects spon- sored by companies. Box (2006) and Zuckweiler (2011) provided an overview of undergraduate quality management courses that incorporated projects to re- inforce basic LSS concepts. Weinstein, Castellano, Petrick, and Vokurka (2008) proposed implementing industry LSS projects to student teams in a MBA quality class. Kanigolla, Cudney, and Corns (2013) analyzed a Six Sigma course that in- cluded industry projects. In this type of student project, firms submit projects (often through active solicitation) and expect student groups to develop results that benefit the company. The PBL pedagogical aspect is apparent in the open-ended nature of the projects that begin with general problem descriptions and unknown solutions.
Several examples of fictional LSS case studies exist. Rasis, Gitlow, and Popovich (2002–2003a,b) described, in two consecutive articles, a fictional Green Belt (person who employees LSS in the context of their normal job) case study suitable for small groups, involving performance and cost improvements associated
384 Lean Six Sigma to the Supply Chain
with paperclips in an office environment. While the study was targeted at Green Belts, it emphasized advanced analytical tools (e.g., Design of Experiment, Mea- surement System Analysis). Johnson, Widener, Gitlow, and Popovich (2006a,b) described an analogous, fictional Black Belt (typically a fulltime or advanced LSS practitioner) case study for optimizing the performance of paper helicopters. The study also emphasized advanced tools (e.g., measurement system analysis, design of experiment, ANOVA). Ellis, Goldsby, Bailey, and Oh (2014) described a large- group (approximately 20 per simulation), multiple-run simulation project in which MBA students managed an airplane order processing supply chain and applied the LSS methodology to make supply chain improvements after each run. Each of these project-based applications emphasizes progressive, tactile learning activ- ities requiring considerable preparation and setup to achieve desirable learning outcomes.
The project described in this article provides an alternative to community engagement projects and facilitated simulations, and requires no physical materials or costs to execute. It is supported by a robust data set, created entirely in Microsoft Excel, designed for specific LSS phases and capable of providing multiple project teams with unique problems (minimizing opportunities for plagiarism). Data can be analyzed with basic (e.g., trend charts, Pareto charts), or more complex (e.g., control charts, capability analysis) LSS analytical tools, providing flexibility for instructors in courses with less quantitative focus. Supporting materials provide sufficient data and guidance to begin each LSS phase, but leave enough ambiguity that students must conduct their own research (using internet and other freely available resources) and demonstrate understanding of LSS concepts and tools to complete the phase objectives. The project topic, cookie baking, was chosen so that students can focus their efforts on the methodology and application of tools rather than getting hung up on technical aspects of the process, and use their own experiences in the kitchen as a source of subject matter expertise. The project detailed below describes a scenario involving a single key process input variable (KPIV) that experienced an undesirable mean shift in one factory location. The prepared data sets provide for similar scenarios involving each of three additional KPIVs, with either a mean-shift or variation-increase issue, at each of two locations. It is thus possible to quickly prepare up to 16 unique project scenarios for groups of 4-6 students. Student instructions and guidance are documented in Appendix A. Instructor guidance is described in Appendix B.
CSU COOKIE COMPANY BACKGROUND
CSU Cookie Company is a manufacturer of premium cookies. Its signature product is the Premium Cookie, a four-inch chocolate treat with chocolate chips in every bite. The Premium Cookie supply chain consists of various domestic ingredients suppliers, two baking locations (Old Facility and New Facility), and a distributor who supplies product to several retail customers. Premium Cookies are known for their high quality, great taste, and low price ($2.00 per package). Annual sales for the product line exceeds $32 million in revenue and 16 million units sold.
A general manager oversees the two identically-outfitted facilities. Each fa- cility has a plant manager, several business unit managers (including the process
Miller, Hill, and Miller 385
Figure 1: Recent financial and quality performance for the Premium Cookie product line.
owner for the Premium Cookie baking unit), shift supervisors, and hourly workers. Products run on a single, dedicated baking process line, with no product comin- gling. Volumes and revenues at both facilities have historically been equivalent, and customers only know which facility made their product by tracking lot numbers, which is rarely necessary. Wages are low, resulting in high turnover and difficulty maintaining skilled labor. Recent cost-reduction efforts resulted in a decision to decentralize ingredient suppliers for key ingredients like flour and sugar to max- imize community goodwill and minimize transportation and storage costs. Each facility reports aggregate performance numbers on a per-shift basis (two shifts per day, 5 days per week) to the general manager. Figure 1 shows the past two quarters of financial and quality performance for the Premium Cookie line.
The general manager is concerned that Premium Cookie revenue has been in decline in the past two quarters, while complaints have increased. Other per- formance measures (e.g., product cost, labor hours, volume, and revenue for other product lines) have not changed, and it is clear that complaints are driving the revenue decrease. She has asked her quality department to form a LSS Green Belt team to investigate and resolve the complaints problem within 90 days.
Scoping the Project
The Green Belt team begins by gathering available business data and creating a high-level process pictorial. A SIPOC diagram (Figure 2) identifies suppliers and customers of each major internal process step, from receiving ingredients to ship- ping finished goods. It also documents major process inputs and outputs. Together, the SIPOC establishes the large boundary for the project; anything outside of the SIPOC is out of scope.
386 Lean Six Sigma to the Supply Chain
Figure 2: SIPOC diagram for the Premium Cookie product.
Figure 3: Old and New Facility Premium Cookie complaints over 6 months.
Since data are reported in aggregate by facility and shift, a pivot table summary stratified by facility and summarized by week (depicted graphically in Figure 3) reveals that the primary problem with complaints originates in the Old Facility. The team is able to narrow the project scope to improvements in the Old Facility, using the New Facility as an internal benchmark.
Closer analysis reveals that the Old Facility complaint level was consistent with that of the New Facility until mid-November, after which complaints began
Miller, Hill, and Miller 387
Figure 4: Pareto chart of complaints at the Old Facility over the past 2 weeks.
increasing to a new plateau level in late December. The team can verify this change with a control chart analysis, but a graphical analysis is sufficient to reduce the project scope further by focusing on changes that occurred in the November timeframe, and using the most recent 3-month window of data to establish process and business baselines. A reasonable operational goal for the project is identified as reducing Old Facility complaints from the current average of 54.5 per shift to the New Facility level of 23.2 per shift (eliminating 15,600 potential complaints per year). An analogous graphical analysis of revenue data (not depicted) reveals a similar pattern, with revenues decreasing through December to a new, lower plateau. A reasonable financial goal is to increase revenue to the level of the New Facility, recovering $57,000 per year in sales. Associated cost-of-poor-quality costs (e.g., returns) will be evaluated in the Control Phase.
By focusing on complaints from the Old Facility, the team can identify the most likely key stakeholder groups and conduct a stakeholder analysis. The general manager (project sponsor) is clearly in support of the project as is the plant manager. However, the owner of the Old Facility baking process is hesitant to have an outside team involved. Process workers are glad to receive the additional help. With this information, the team can plan a strategy to capitalize on the supportive stakeholders and win over the support of the process owner. The last key stakeholder group, the customers, requires more in-depth analysis.
Analyzing Voice of the Customer
Getting complaint data from the company database is a challenge, but the most recent 2 weeks are readily available. A Pareto analysis (Figure 4) quickly reveals that the vast majority (70%) of recent complaints involve taste. The team then con- ducts a critical-to-quality (CTQ) analysis to further understand how the customer defines “taste.”
CSU Cookie Company measures taste for each cookie batch using a propri- etary taste technique and professional taste testers. The measure is well-calibrated,
388 Lean Six Sigma to the Supply Chain
Figure 5: CTQ tree defining customer need for quality cookie in terms of process CTQs.
and a score of 90 or above (out of 100) is minimally acceptable. After interviewing sales people who represent the customer (simulated by internet search or inter- viewing classmates for how they define “good tasting cookie”), the team is able to use an affinity technique to group their feedback by common themes and derive a definition for taste in terms of texture, firmness, and chocolate flavor. Each of these drivers can be defined in terms of process variables with measures (see Figure 5 for CTQ tree), and the process owner confirms that each measure has defined specifications the process team will provide later.
With the CTQ analysis, the team can further narrow the project scope to complaints in the Old Facility associated with unacceptable taste. They are ready to prepare a project charter.
Chartering the Project
The project team has enough information to complete a project charter (Figure 6), including problem statement, business case, operational and financial impact, project goals, and resources.
To complete the Define Phase, the team prepares two basic project man- agement tools: a project communication plan for key project stakeholders, and a Gantt chart depicting a detailed timeline for each project phase, with the phase milestones included in the charter document. The team documents its work and presents the Define Phase tollgate to the project sponsor for approval.
Documenting the Current-State Process
The CTQ tree provides a reasonable start for baseline process analysis, identifying the key process output variable (KPOV) as taste, and the KPIVs as bake time, bake temperature, cookie dough viscosity, and cookie dough weight; this makes it possible to prepare a data collection plan (DCP). In addition, team members employ
Miller, Hill, and Miller 389
Figure 6: Project charter for CSU Cookie Company complaints reduction project.
video process mapping (simulated by available YouTube videos, see Appendix B) to prepare a process map (PMAP) (Figure 7) with enough detail to contain the process steps that involve these key variables.
It is likely that root causes of any complaint will reside somewhere within this detailed process. If the investigation does not reveal root causes, the team will have to retrace their investigative steps and potentially expand their process picture. To avoid this outcome, the team will analyze the entire baking process from ingredient preparation to packaging. In the course of preparing the PMAP, the team identifies that key process variable data is captured manually, reported as averages by shift, and that underlying raw data are not saved. Process data are stored in a local database to which the process owner grants them access. It contains two quarters of data for the KPOV and each KPIV, and provides process specifications for each variable.
Establishing the Current-State Process Baseline
With process data provided by the process owner, the team is able to analyze baseline performance using individuals control charts. For the input variables of bake time (target = 10 ± .25 minutes) and bake temperature (target = 350 ± 5 °F), control charts reveal that behavior is stable and in control, process means are properly at target values, and process variables are capable (control limits are inside specification limits). For taste (KPOV) and dough viscosity (KPIV), the story is different (see Figure 8).
The KPOV control chart reveals a process that is in control, but with a mean of 85.1 (target of 95 ± 5 units) and incapable of meeting customer specifications.
390 Lean Six Sigma to the Supply Chain
Figure 7: Current-state detailed process map for Old Facility baking process.
Figure 8: Control charts for underperforming KPOV (Taste) and KPIV (Dough Viscosity), with customer specification limits for capability comparison.
The KPIV chart reveals a process that is likewise in control, but incapable of meeting customer specifications (viscosity target = 50 ± 5 centipoise [cP]). It is clear from the analysis that variation for each variable is acceptable but that means must be shifted to improve the process.
Capturing Potential Causes
The team begins looking for potential causes of poor taste prior to completing their quantitative baseline analysis. They interview subject matter experts on the baking
Miller, Hill, and Miller 391
Figure 9: Fishbone diagram with potential causes for poor-tasting cookies.
process (simulated by internet searches or interviewing classmates who like to cook) and brainstorm potential causes based on the current baking process. Results are summarized in a fishbone diagram (Figure 9). Results from the quantitative analysis will narrow their Analyze Phase search for root causes. The team prepares a Measure Phase tollgate report for approval by the project sponsor.
Identifying the Root Cause
From the baseline process analysis, the Green Belt team knows that the process input cookie dough viscosity, resulting from the compounding and mixing of raw material ingredients, experienced a shift away from the target mean concurrent with a decrease in taste. From the fishbone diagram they are able to eliminate potential causes of poor taste related to other process inputs, and focus on what is causing the poor viscosity performance. Using the detailed process map, they prepare a Failure Modes and Effects Analysis (FMEA, see Figure 10) to correlate potential causes with process steps, then rate potential causes and effects using a Likert Scale (1 is always good or low, 10 is always bad or high) on the basis of frequency of occurrence, severity of effect, and effectiveness of current process controls. (Teams are given guidance on how to rate potential causes based on their experience with baking or with internet research, but they are not graded on the specific causes that rate the highest.)
Multiplying the three ratings produces a risk priority number (RPN) for each potential cause, allowing the team to identify two key contributors with the highest RPNs, poor-quality flour, and improper mixing execution.
392 Lean Six Sigma to the Supply Chain
Figure 10: FMEA for current-state baking process.
Process Step Failure Mode Failure Effects Sev Causes Occ Controls Det RPN
Sourcing Poor flour quality mix too thin 5 New supplier 10 None 10 500
Mix Ingredients Incorrect mixing order Too soft or too hard 7 Operator not following SOP
9 Training 6 378
Mix Ingredients Bad flour Dough too thick or thin
7 Flour does not meet specs
9 Vendor BOM 6 378
Mix Ingredients Incorrect mixing order Unpleasant taste 7 Operator not following SOP
9 Training 6 378
Package cookie Unsealed container Too hard 8 Low seal pressure 6 Monthly preventive maintenance
Package cookie Air in bag before sealed Too hard 6 Suction hose clamped/clogged
8 Monthly preventive maintenance
Gather Ingredients Incorrect flour qty Too soft or too hard 5 Hopper feeding incorrectly
7 Monthly preventive maintenance
Package cookie Air in bag before sealed Too hard 6 Suction hose came loose
4 Monthly preventive maintenance
Package cookie Air in bag before sealed Too hard 6 Timing off- sealing before suction
3 Monthly preventive maintenance
Mix Ingredients Incorrect mixing rate Too soft or hard 7 Incorrect SOP 3 Quarterly SOP review 3 63 Mix Ingredients Incorrect mixing duration Unpleasant taste 7 Incorrect SOP 3 Quarterly SOP review 3 63 Gather Ingredients Incorrect flour qty Too soft or too hard 5 Scale not calibrated 5 Monthly calibrations 2 50
Sourcing Incorrect flour mix Too hard 5 New supplier 1 None 10 50
Package cookie Unsealed container Too hard 8 Packaging material wet
1 Incoming QA check 1 8
Figure 11: Box plot comparing current process to test for root causes.
Verifying Root Cause
Multiple observations of mixing operations confirm that to save time mixing oper- ators have developed their own short-cuts to the standard operating procedure. The lead baker verifies that mixing the dough at a higher rate may save time but that too much shear stress causes the dough to become runny (lowering viscosity). To test the flour source, the team prepares a simple test using resources in the plant’s quality laboratory. Operators prepare several batches of Premium Cookie in the lab using excess flour from the previous flour vendor, and compare results to random samples taken from the production line (Figure 11).
Miller, Hill, and Miller 393
Figure 12: Pugh matrix comparing potential improvements.
A box plot of the results shows that dough viscosity using the old flour source is within specifications, and the plant taster agrees that the cookies are delicious. A 2-sample t test confirms a significant difference in mean viscosity between sample groups (p = .000). The team presents its findings to the process owner and project sponsor in the form of an Analyze Phase tollgate.
Selecting the Solution
Brainstorming potential solutions results in several worth considering, so the Green Belt team prepares a Pugh matrix to compare alternatives on the basis of effort and impact (Figure 12).
The process owner agrees that the appropriate solution will involve using the previous flour vendor while properly qualifying the new vendor, and implementing a long-term training program for all process operators.
Verifying the Solution
Before committing to the process changes, the team develops a pilot plan for a 30- day trial period. The old flour vendor agrees to provide flour at a discount during this time (they want the business back), and the training team agrees to mentor all mixing operators during their regular shifts to ensure compliance with SOP. All process variables are monitored closely by the Green Belt team for changes. Results are immediately clear: dough viscosity falls back into specifications and
394 Lean Six Sigma to the Supply Chain
Figure 13: Control charts for improved KPOV (Taste) and KPIV (Dough Viscos- ity), with customer specification limits for capability comparison.
taste improves to acceptable levels (see Figure 13 for control charts of viscosity and taste). An updated (post-RPN) FMEA shows expected improvements based on lower RPNs.
Performance for other KPIVs remain unchanged. Toward the end of the trial period, the team prepares the Improve Tollgate to share its results.
Implementing and Controlling Improvements
The plant manager is not willing to permanently align trainers with mixing opera- tors, and the Green Belt team does not want to permanently monitor the process. They prepare an implementation plan (Figure 14) that includes training all current employees, visual SOPs that are laminated and posted by the mixing stations, updates to the new-hire SOP to include trainer mentoring at all process stations as part of onboarding training, and purchasing changes to re-engage the old flour vendor. Purchasing agrees that properly sourcing new vendors is a significant ef- fort that they will address in a separate project. A process control plan (Figure 14) emphasizes task ownership at the operator level where possible, including process monitoring and an action plan when variables misbehave.
Closing the Project
With the improvements implemented, results verified, and control plan handed off to the process owner, the team summarizes project benefits and presents the final Control tollgate to the key stakeholders (see Figure 15 for an executive summary).
Final improvements include a 57% reduction in complaints for the most- recent reporting period, and a revenue increase that amounts to an annual gain of more than $57,000, both results that meet project goals. The plant controller confirms that a decrease in costs associated with returns translates into an additional $31,000 in annual savings. The project sponsor is happy.
Miller, Hill, and Miller 395
Figure 14: Control plan for improved baking process.
EVIDENCE OF EFFECTIVENESS
During spring 2015, the project was used in an undergraduate course on supply- chain financial issues and in two sections of a graduate course on supply-chain field problems as a self-directed project with no formal class time devoted to LSS lectures. A twenty question pretest representing each of the five LSS phases was administered prior to beginning of the project. Fifteen multiple-choice questions, three for each LSS project phase, were included in student grades. Five additional questions were used to identify future assessment revisions. The same instrument was administered as a posttest to assess student learning. Overall, the pretest mean (N = 45) was 41.5%, and the posttest mean was 60.1%, a nominal increase of 18.6% percentage points and a 45% increase from pretest baseline. Results for graduate students (41.4% pretest, 62.8% posttest, 51.7% increase from baseline) were slightly higher than those for undergraduate students (41.6% pretest, 57.1% posttest, 37.4% increase from baseline).
In fall 2015, the project was again delivered in an undergraduate course on supply-chain financial issues. A formal overview of LSS was provided prior to the project and a discussion of project progress integrated into lectures. The pretest was expanded to 25 questions (15 questions from the spring quiz, plus an additional 10) to better assess project activities in each phase. Pretest scores (51.7%, N = 12) were higher, which we attribute to the formal overview. Posttest results (77.3%) were higher, too, which we attribute to reinforcement throughout the term. Informal student feedback indicated that the project was well received, and formal feedback included course comments including “The project was the highlight of this course” and “The Six Sigma project was very eye opening and beneficial.”
396 Lean Six Sigma to the Supply Chain
Figure 15: Control Phase executive summary with documented improvements and benefits.
The LSS project provides faculty with an alternative to community engagement projects and facilitated simulations to teach and reinforce LSS concepts and tools using a simulated business problem. Its successful deployment in both undergrad- uate and graduate supply chain courses supports its utility. Its ease of setup, limited active facilitation requirements, elimination of physical materials and associated costs, and data set flexibility, make it scalable and adaptable in a variety of course delivery and schedule formats. The use of a familiar project topic, cookie baking, makes the project both approachable and relatable for a wide range of students. And, while the problem is sufficiently ambiguous to require students to critically evaluate options and apply tools to derive unique solutions, the available robust dataset allows faculty to control the outcomes enough to ensure the consistency and quality of learning. In the future, the project will be further refined and will result in the development of additional data sets and company scenarios, and its implementation tested in an online course environment.
Anderson-Cook, C. M., Patterson, A., & Hoerl, R. (2005). A structured problem- solving course for graduate students: Exposing students to six sigma as part of their university training. Quality and Reliability Engineering International, 21, 249–256.
Miller, Hill, and Miller 397
Box, T. M. (2006). Six Sigma quality: Experiential learning. Sam Advanced Man- agement Journal, 71(1), 20–23.
Breyfogle III, F.W. (2008). The inside track on six sigma training. Quality, 47(2), 46–52.
Duch, B. J., Groh, S. E., & Allen, D. E. (2001). Why problem-based learning? A case study of institutional change in undergraduate education. In B. Duch, S. Groh, & D. Allen (Eds.), The power of problem-based learning. Sterling, VA: Stylus, 3–11.
Ellis, S. C., Goldsby, T. J., Bailey, A. M., & Oh, J. Y. (2014). Teaching lean six sigma with a supply chain context: The airplane supply chain sim- ulation. Decision Sciences Journal of Innovative Education, 12(4), 287– 319.
Genis, V., Mauk, M., Gogotsi, Y., Sakalley, D., Hagarman, J., & Burnside, H. (2010). Lean Six Sigma nanomanufacturing course for undergraduate en- gineering technology and engineering programs. Journal of Engineering Technology, 27(2), 18–29.
Goldsby, T., & Martichenko, R. (2005). Lean six sigma logistics: Strategic devel- opment to operational success. Florida: J. Ross Publishing, Inc.
Hmelo-Silver, C.E. (2004). Problem-based learning: What and how do students learn? Educational Psychology Review, 16(3), 235–266.
Johnson, J. A., Widener, S, Gitlow, H., & Popovich, E. (2006a). A “Six Sigma” C© black belt case study: G.E.P. Box’s paper helicopter experiment part A. Quality Engineering, 18, 413–430.
Johnson, J. A., Widener, S, Gitlow, H., & Popovich, E. (2006b). A “Six Sigma” C© black belt case study: G.E.P. Box’s paper helicopter experiment part B. Quality Engineering, 18, 431–442.
Kanigolla, D., Cudney, E. A., & Corns, S. M. (2013). Employing project-based learning in six sigma education. The Journal for Quality & Participation, 36(1), 24–38.
Pyzdek, T., & Keller, P.A. (2014). The six sigma handbook (4th ed.). New York, NY: McGraw-Hill.
Rasis, D., Gitlow, H. S., & Popovich, E. (2002–2003). Paper organizers interna- tional: A fictitious six sigma green belt case study. I. Quality Engineering, 15(1), 127–145.
Rasis, D., Gitlow, H. S., & Popovich, E. (2002–2003b). Paper organizers interna- tional: A fictitious six sigma green belt case study. II. Quality Engineering, 15(2), 259–274.
Schafer, S. M. (2005). Karate in business school? This is not your father’s black belt. The Quality Management Journal, 12(2), 47–56.
Weinstein, L. B., Castellano, J., Petrick, J., & Vokurka, R. J. (2008). Integrating Six Sigma concepts in an MBA quality management class. Journal of Education for Business, 83(4), 233–238.
398 Lean Six Sigma to the Supply Chain
Zuckweiler, K. M. (2011). Teaching six sigma to undergrads: A simplified real project approach. Decision Sciences Journal of Innovative Education, 9(1), 137–142.
APPENDIX A: LSS GROUP PROJECT INSTRUCTIONS TO STUDENTS
CSU Cookie Company is a bakery supplying premium-quality cookies to retailers throughout the Southeast United States. An overview of your value stream is below:
You operate two facilities, each producing the same product mix to meet de- mand. One facility is original to your company, and the second is new, becoming operational in the past year. Information for your bakery is controlled centrally for monitoring and reporting purposes, and distributed to your factories. In the past two business quarters, you have seen a small but steady increase in customer complaints, and a corresponding decrease in revenue from your main cookie prod- uct. Volumes and revenues at both facilities have historically been equivalent, and customers only know which facility made their product by tracking lot numbers, which is rarely necessary. Wages are low, resulting in high turnover and difficulty maintaining skilled labor. Recent cost-reduction efforts resulted in a decision to de- centralize ingredient suppliers for key ingredients like flour and sugar, to maximize community goodwill and minimize transportation and storage costs.
Your team is tasked with executing a process improvement project to identify and address the issues resulting in increased customer complaints and decreased revenue. The leadership team would like to see demonstrated improvements within 2 months.
Miller, Hill, and Miller 399
Project Status Reporting
Regular status reports (updated timeline, progress, next steps, and issues) should be submitted in accordance with the published schedule (see course syllabus), using the provided MS Excel format.
Data Set Availability
Data sets specific to each phase are provided as follows:
1. Business baseline data (Define Phase) upon team selection.
2. Process baseline data (Measure Phase) upon completion of Define Phase.
3. Test data (Analyze Phase) upon completion of Measure Phase.
4. Process postimprovement data (Improve Phase) upon completion of Analyze Phase.
5. Business postimprovement data (Control Phase) upon completion of Improve Phase.
Graduate teams may provide their own business problem and data, providing at least one team member can serve as a process subject matter expert. Any data or other information used must be cleared by appropriate business authorities for public sharing.
1. Status reports: Submitted in accordance with Course Syllabus due dates (one per phase, mid-way to tollgate due date).
2. Process improvement tools: Suggested LSS tools are listed in the table below. Each team member must complete, at a minimum, one required tool from each phase, and all tools listed as required must be completed by the team. Optional tools for large teams must be approved by the professor in advance, so account for this requirement in your project planning. You may help each other with tools, but each member is responsible for completing their own tool for each phase.
3. Project tollgates (phase-based reports): PowerPoint format (provided in class), including (1) executive summary slide, (2) detailed tool output with key points on slides, (3) additional details (including purpose for choosing tool, and name of responsible student) in speaker notes, and (4) table indicating team member contributions. Each tollgate must contain the previous tollgate(s) with all instructor feedback addressed. The final (Control) report will contain all five project phases in order, Define first and Control last.
4. Group presentation: 15-minute class presentation depicting the pro- gression of the team’s project work from Define through Control. Each team member must participate in the group presentation.
400 Lean Six Sigma to the Supply Chain
Required and recommended project tools by phase.
Deliverables by Project Phase
Helpful Tools Aligned with Objectives
Case Study Suggested Use
1. Project charter Charter template Required 2. High-level process
process map Required
Value stream map (VSM)
3. List of customer(s) and project CTQs (critical to quality)
Voice of the customer (VOC) ranking matrix
VOC data collection plan Optional Graphical analysis
(Pareto, etc.) Optional (as data is
available) Affinity diagram Optional CTQ tree Required
4. Communication plan, project-focused
Stakeholder analysis Optional Stakeholder
communication plan Required (large group)
6. Project plan/timeline Gantt chart Required 7. Project benefits Operational analysis Required (large group)
Cost-benefits analysis Required Measure Phase
1. Detail the process Detailed process map/flow chart
VSM (if necessary and not done in Define Phase)
Nonvalue add (NVA) analysis (Measure or Analyze Phase, not both)
2. Define process outputs (Y) and potential drivers (x’s)
Ishikawa (Fishbone) diagram
Cause-and-effect matrix (often completed in Analyze Phase)
3. Define defect(s) and set specification limit(s)
Defect definition table Optional
4. Data collection plan Data collection plan table
Sample calculator (Continuous & Discrete Data)
Miller, Hill, and Miller 401
Deliverables by Project Phase
Helpful Tools Aligned with Objectives
Case Study Suggested Use
5. Validate the measurement system
Measurement system analysis (MSA)
Out of scope for assignment
6. As-is process capability
KPOV/KPIV control charts or other analytic technique
Process capability baseline
7. Update project benefits
Updated cost-benefit analysis
1. Verify cause and effect relationship between critical inputs and critical output variables
Cause-and-effect matrix (If not done in Measure Phase)
FMEA or Murphy analysis
Graphical analysis Required (at least one graph)Box plots
Histograms Pareto charts Scatter plots Statistical tools (t-test,
ANOVA, etc.) Required (at least one
statistical test) Hypothesis testing
2. List of verified root causes
5 Whys Optional Brainstorming Optional List of significant
inputs/root causes Required
1. List of improvements Brainstorming Optional Pugh matrix or solution
selection matrix Required
2. Identify quick hits Quick wins Optional Rapid improvement
events/Kaizen Out of scope for
assignment 3. Validated
improvements pilot plan Required (large group) Control chart/graphical
analysis Required (at least one
graph) Postimprovement FMEA Required
4. Improved process documentation
Future state process map Required for process change
402 Lean Six Sigma to the Supply Chain
Deliverables by Project Phase
Helpful Tools Aligned with Objectives
Case Study Suggested Use
1. Process control plan KPIV/KPOV control chart/graphical analysis of pre- and postimprovements
Control plan Required Standard operating
procedures (SOP) Optional
2. Communication plan, process focused
Communications plan Optional
3. Implementation plan Implementation plan Required 4. Visual process control
tools Primary metric chart Required Precontrol chart Optional
5. Final project benefits Cost-benefit analysis Required
GUIDANCE TO STUDENTS BY PHASE
Each of the following blocks of information are provided to students at the begin- ning of each project phase. Unless otherwise noted, the following guidance can be included in each phase:
Resources available: In addition to tool templates provided in class, you can find tool templates, instructions, and examples on YouTube, iSixSigma.com, or through a targeted internet search. You must cite any online resources used in the speaker notes section of your project report.
Data available: Each phase incorporates a data file provided by the instructor (see each phase for details). For each phase, the instructor may also provide additional data you may need, provided you have a reasonable data collection plan and allow for a 1-week turn-around.
Define Phase Background and Guidance: “What Is the Problem?”
You will define the boundaries of your investigation and the scope of your work. Analyzing available business data may help to determine whether the problem is systemic for the product, or specific to a location. Narrowing the scope of work simplifies future efforts and maximizes available resources. If the problem appears most prominent in one location, include that location in the scope of work and exclude other locations. For example, comparing revenue and volume data between the two manufacturing sites may help you to narrow your focus to one location, using the other location as an internal benchmark. A high-level pictorial, such as a SIPOC, helps to establish boundaries, manage scope-creep, and determine key stakeholders. Available complaint data will help focus your voice of the customer analysis. Once you determine the biggest complaint area, you can narrow your project scope to that category, saving other complaint areas for future projects. Your analyses should culminate in a project charter with a quantitative project goal, and a deliverable-based project timeline.
Miller, Hill, and Miller 403
Data available: An Excel file is provided, containing two quarters of business data, including revenue, volume, and cost across both manufacturing locations, and complaint data for the most recent 2-week period. The data are presented as summaries per shift, with two shifts per day and five working days per week. You will want to use pivot tables or otherwise summarize your data by day or week to simplify your analysis.
Additional Resources available: The instructor will serve as a primary source of data from any stakeholder groups you identify, providing you come prepared with a data collection plan and focused questions. However, you can simulate stakeholders such as customers and process owners using your team members, peers, and others who may have ideas on what could be important to the business and to the customer. The instructor will serve as your Master Black Belt coach throughout the project provided you come prepared with your work products and specific questions related directly to your project phase.
Measure Phase Background and Guidance: “How Big Is the Problem?”
You will conduct a current-state (baseline) process analysis. A detailed process map should focus on the area of the supply chain (depicted in your SIPOC) that most likely contains the problem, in this case, the cookie production process. Available key process output variable (KPOV) and KPIV data should be analyzed for process stability and conformance to the voice of the customer (represented by variable specifications). Properly performing KPIVs may be excluded from the project scope, and underperforming variables should serve as your problem-solving focus moving forward. Potential causes for underperforming process variables can be captured using brainstorming in conjunction with cause-and-effect tools.
Data available: An Excel file is provided containing two quarters of process data by location. The instructor will provide any additional data you may need provided you have a reasonable data collection plan and allow for a 1-week turn- around. This company uses a proprietary taste test for its KPOV. Each location also monitors the following KPIVs:
� Viscosity: the consistency of cookie dough, indicating proper ingredient preparation
� Cookie weight: the prescribed quantity of dough for each cookie � Bake time and bake temperature: baking process conditions
Resources available: The assignment focuses more on the ability to use indi- vidual analytical tools correctly and tie them together into a problem-solving story, and less on the technical accuracy of an actual manufacturing plant. Some basic tool templates are provided in class, including an Excel control chart generator and an Excel fishbone diagram template. For graphical and statistical analysis, you may use Excel functions or search for freely-available templates on the internet. For subject matter expertise, you can, first and foremost, use your own experience with baking as the basis for subject matter expertise. Other potential resources include
404 Lean Six Sigma to the Supply Chain
� A YouTube search for “cookie manufacturing” or additional targeted in- ternet searches to simulate “walking the process” and developing your detailed process maps
� Prepare and bake your own cookies to document your detailed process map
� Brainstorm using your collective experiences and those of your peers
You will be evaluated on your ability to employ the problem-solving tools in conjunction with the methodology and not on the specific content of your potential- cause brainstorming. However, you should be able to rationalize a connection between any potential causes you list for the problem you identify.
Analyze Phase Background and Guidance: “What Is Causing the Problem?”
You will narrow down your list of potential causes to a short list of verified, actionable (items within the control of your project team that can be fixed) root causes. This phase requires some creativity; the focus should be on the process of discovery and verification. Use your own subject-matter expertise and that of your peers, internet searches, and (as a last resort) your Master Black Belt coach to prioritize the list of potential causes. Tools such as FMEA and cause-and-effect matrix are useful. It does not matter so much which potential causes you assigned highest priority as long as you can explain your logic. For example in an FMEA, a potential cause that results in an inconsequential defect should have a lower severity than one that could result in injury. Your team should pick 1-3 potential causes that have the highest priority, and for this exercise you can declare them to be the root cause, understanding that in a real project you would need to collect more data to verify your conclusion. To simulate this verification, you should describe a simple test that you could conduct that compares current-state process performance to some simulated condition with your root cause removed. You will be provided appropriate sample data to conduct a statistical test verifying the root cause(s).
Data available: An Excel file is provided with a sample set containing 20 experimental data and 20 control data for a root cause verification statistical test.
Improve Phase Background and Guidance: “How Do We Fix the Problem?”
You need to conceive, compare, and test improvements that will reduce or eliminate your identified root causes. You should identify at least three courses of action that could reasonably address the root cause, and choose one course of action to evaluate with a pilot plan. Analysis of postimprovement process data should be similar to your analysis from the Measure Phase. A comparative analysis (graphical or statistical) should demonstrate a sustainable improvement to the KPIV and KPOV.
Miller, Hill, and Miller 405
Data available: An Excel file is provided with 30 days of postimprovement process data by location. You can combine this data with your baseline process data to conduct any necessary comparative analysis.
Control Phase Background and Guidance: “How Do We Maintain the Improvements?”
You need to document a solution implementation plan, control plan you can hand off to the process owner, and updates to any standard operating proce- dures or other process documents. A final operational analysis (using the key business operational metric of complaints) and financial analysis should docu- ment your project’s outcomes relative to project goals. The final project report should contain all five project phases from Define through Control, with appro- priate executive summaries, and other annotations required in the assignment instructions. This PowerPoint deck should serve as the foundation of your team presentation.
Data available: An Excel file is provided with all pre- and postimprove- ment data including 30 days of postimprovement business data by location. You should not need any additional data to complete this phase and close the project.
APPENDIX B: GUIDANCE FOR INSTRUCTORS
This project represents a compromise between cases that involve considerable preparation and activities that are designed for full-time employees working to- gether in a dedicated LSS training environment, and textbook exercises that demon- strate applications of individual tools but are not necessarily progressive applica- tions of tools to resolve a complex problem. There are several predefined outcomes embedded throughout the project that allow instructors to effectively control the learning environment. However, students must follow the LSS methodology to arrive at those outcomes. For example, the overarching scenario begins with a manager worried about rising complaints, and students must follow the method- ology to define an actionable problem and scope the project for their available resources and data. Each current-state process and subsequent improvement have similar types of issues and magnitudes of improvement. Students must use a col- lection of problem-solving tools to determine the specifics for their case. The project may be deployed as a whole, or each phase may be used as individual or small-group assignments throughout the term.
Data sets are segmented by facility (new facility and old facility), underper- forming input variable (bake time, bake temperature, cookie weight, and cookie dough viscosity), and process issue (variation issue and process-shift issue), for a total of 16 sets that accommodate small or large classes or multiple class sections. All data sets are designed to result in similar (but not identical) quantitative project goals and outcomes.
406 Lean Six Sigma to the Supply Chain
After completing this project, students will be able to
� Describe the phases of the Lean Six Sigma process improvement methodology.
� Describe and apply problem-solving tools in support of Lean Six Sigma phase-based project objectives.
� Apply the Lean Six Sigma methodology and appropriate supporting tools to the critical evaluation and resolution of a business problem involving underperforming process input and output variables.
ANALYSIS BY PHASE
There are five critical phases to the project, each of which builds on the previous phase for a progressive understanding of, and resolution to, a simulated business problem. The core data sets provide the instructor with a high degree of consis- tency and control between project teams without compromising the uniqueness of approaches and solutions. Students should be evaluated on the correct applica- tion of tools to reach objectives, and answer the critical question for each phase. Analytical tools should be evaluated based on outcomes derived from the data provided; qualitative tools should be evaluated based on the correct application of the tool and integration into the story evolution, and not on the specific items that students conceive for potential and root causes.
Students typically struggle to get started, focusing on the completion of individual tools that are required rather than on meeting phase objectives. Feedback on the Define Phase tollgate should focus on the interconnectedness of the tools as much as on the technical correctness of each tool. For example, the VOC analysis should be supported by the complaints analysis, and charter goals should come from the financial and operational analysis. Requiring an instructor-led “consulting session” part-way into the Measure Phase helps students to visualize how the tools relate to one another and come together for the phase objectives. By the Analyze Phase, students are typically applying tools correctly in the context of their project with little guidance, and feedback shifts to details that provide consistency between phases. An additional consulting session part-way into the Improve Phase is often enough for students to close out their projects with confidence.
Comparison of data from the two manufacturing locations leads to the identification of one underperforming location which should remain in scope for the project duration. The well-performing facility should be out of scope for improvements, but should serve as a benchmark for quantitative operational and financial project goals; the goal should be to get the underperforming facility performance (based on complaints, the key business operational metric, and Revenue, the key business financial metric) back in line with the other location. Students are provided with suggestions for online resources to use as background for cookie manufacturing; YouTube provides numerous short documentaries that students may use for SIPOCs
Miller, Hill, and Miller 407
and detailed process maps, or students may serve as their own subject matter experts based on their own cooking experience.
Once students resolve where the problem resides and complete a high-level process pictorial, they should be able to identify likely key stakeholders including the process owner, process workers, and the business manager (project sponsor). Consumers should be identified as a key stakeholder for voice-of-the-customer (VOC) analysis, but not necessarily for intra-project communications or stake- holder analysis. Analysis of complaints data should lead to the conclusion that poor taste is the issue, which will simplify VOC analysis. A simple project plan in the form of a Gantt chart based on phase-based milestones should be a requirement. A project status report should be submitted roughly half-way to completion for this (and every) phase. We provide a simple, one-page status report template that incorporates a Gantt chart and allows the instructor to quickly determine targeted guidance for each team.
The Measure Phase is easiest to manage for the instructor and students if it is di- vided into three components, pictorial analysis, quantitative analysis, and potential causes. For the pictorial analysis, the details of the detailed process map are not as important as the technical correctness of creating the flow diagram and capturing all steps in the process that encompass process inputs and outputs. In this case, the map should cover ingredient preparation (mixing) to finished cookies. Personal experience or internet videos will suffice for the level of subject matter expertise required. Quantitative analysis should quickly reveal that the output variable (taste) is underperforming, and that one associated input variable is underperforming. All other input variables are in control and capable. The data sets are designed to require either time-based trend charts or simple individuals control charts if that technique is part of the curriculum. No subgrouping or discrete control charts are necessary. Students may try to brainstorm potential causes based on the effect of high complaints, but they should be guided by their analysis to narrow the focus to the process variables. Students may choose to brainstorm causes associated with the underperforming input or output variable. While capability analysis is feasible for this project with the data sets provided, we choose to emphasis only qualitative comparison of control limits or raw data to specification limits to visu- alize whether a variable is meeting specifications. This emphasis is based on our industry experience that leaders respond to graphics and rarely understand capabil- ity calculations. Measurement System Analysis is not emphasized in this project. However, measurement-related issues including inadequate data collection plans, poor or missing operational definitions, or unreliable contextual data, may be used to foster classroom discussion around potential causes and potential ways verify causes, and around measurement system evaluations based on discrete (e.g., taste) or continuous (e.g., temperature) data.
The Analyze Phase is the least quantitative, and requires the most creativity, of any phase in the project. By this point the student teams should have a sound baseline
408 Lean Six Sigma to the Supply Chain
understanding of the process, and a clear picture of how the LSS methodology is leading them to an actionable root cause for the process issue. There is limited data available in this phase due to time constraints and the desire to manage project complexity. Students typically take their list of potential causes and prioritize them using an FMEA or a cause-and-effect matrix. Which potential causes they choose to prioritize is not important, only that they choose 1-3 and explain in logical terms why those may be the most likely causes. They should prepare a simple test, for example a Quality Assurance lab test, to compare some experimental data to random process data using a basic statistical test with provided sample data. Which statistical test they employ depends on whether they were assigned a variation problem set (2-variances test) or a baseline-shift problem set (2-sample t-test).While a Design of Experiments is not required for this project, the instructor may choose to provide data that allows for fractional or full-factorial analysis in this phase.
At this point in the project, student teams should have a clear mental picture of the end state. How the students choose to address their root cause is not important, as long as the solutions are reasonable for time and resource availability (90 days for the completed project, and no capital funding available). They should conceive of several courses of action and use a criteria-based selection tool to compare and choose one for testing. A pilot plan should focus on testing improvements identified for the root causes and incorporate the available postimprovement pro- cess data. Quantitative analysis for this phase mirrors analyses employed in the Measure Phase using postimprovement data. Students should demonstrate that the underperforming variables are now capable and in control, and that other variables are unchanged. For planning purposes, the Improve Phase should go faster than the prior phases.
In our experience, the Control Phase is the most overlooked LSS phase; teams have demonstrated their improvements and want to close the project. Unfortunately, the long-term effects are typically a business unit inundated with countless process control plans, all to be managed by a process owner. The critical elements for this phase are thus the implementation and control plans, targeted at the lowest level of reasonable supervision. The before-after business analysis should be targeted at the project sponsor, and should demonstrate how close the project team came to reaching the stated project goals.
This project is designed to be completed in groups of 3-6 students each, with the majority of work being performed outside the classroom. We have found that a solid overview of the LSS methodology and a walk-through of a completed LSS case study are sufficient for students to get started. Each student should be required to complete a minimum of one tool per phase, with a minimum of four required tools per team for each phase. This requirement provides each team with a total
Miller, Hill, and Miller 409
of 20-30 tools applied throughout the project. Student teams are typically left to self-select the tools assigned to each member from the list provided in the project instructions. Below is a table of recommended time allocation per project phase, based on a 16-week semester.
LSS Supporting Project Student Classroom Topic Materials Duration Deliverable
Lean Six Sigma Overview
Topic Selection One Week Project Plan
Define Phase Concepts
Define tools and examples
Two Weeks Define Tollgate
Measure Phase concepts
Measure Tools and examples
Three Weeks Measure Tollgate
Analyze Phase concepts
Analyze Tools and examples
Two Weeks Analyze Tollgate
Improve Phase concepts
Improve Tools and examples
Two Weeks Improve Tollgate
Control Phase concepts
Control Tools and examples
Two Weeks Control Tollgate
Delivery timing for data sets is important to simulate real-world data avail- ability, to minimize student analysis paralysis, and to keep groups from working ahead. At the project kickoff, students are provided with baseline business data from two facilities, and accompanying complaints data over a recent 2-week pe- riod. Upon the Define tollgate turn-in, students are provided with process baseline data to conduct their Measure Phase analysis. We provide baseline data for all facilities and process variables, and leave it to the students to manage their project scope by focusing only on the underlying facility. When the Measure tollgate is complete, we provide a small Analyze Phase data set that simulates a simple exper- iment with a control group. From this, students can describe a possible experiment to test for a root cause, and conduct a basic statistical test to verify it. At the be- ginning of the Improve Phase, we provide a complete set of postimprovement data covering a 30-day period. Regardless of the underlying process issue, the improved data set shows a sustainable improvement in the underperforming variables. For the Control Phase, we provide a complete set of postimprovement business data (30 days), including a recent 7-day collection of complaints, for the final analysis.
Each project phase culminates in the tollgate deliverable. It is important to provide feedback on each tollgate early enough in the next project phase that students can make corrections or adjustments and incorporate them into the project. Requiring each tollgate to include revised slides from previous phases provides students with a complete take-home case study at the end of the course. It also minimizes end-of-semester presentation preparation.
To keep project teams moving forward at approximately the same pace, we require at least two, 1-hour “consulting sessions” between instructor and team, either in-person or virtually, depending on the comfort level of the instructor and the team. The first session is required part-way into the Measure Phase to provide feedback on the Define Phase and answer questions about the path forward. The
410 Lean Six Sigma to the Supply Chain
second session is typically required late in the Analyze Phase or early in the Improve Phase for the same reason. Timing works well with the course in that the first session occurs before midterm and the second occurs after midterm.
Group presentations should mirror typical LSS green belt project report-outs, with executive summaries for each phase and selected key tools but not necessarily every tool. Fifteen minutes of presentation and 5 minutes for questions is often sufficient and representative of how much time a real project team might have to present and defend results. To accommodate large numbers of groups and limited time, presentations may be based on project-phase executive summaries only, demonstrating the effectiveness of properly-designed executive summaries.
A completed case study tollgate file (PowerPoint format), and a complete set of data for up to 16 project teams (Excel format) are available from the authors upon email request. There are many LSS texts and handbooks available, including:
� Pyzdek and Keller, Six Sigma Handbook, Fourth Edition. McGraw-Hill: Chicago.
� Brassard, Finn, Ginn, Ritter, Kingery, and Kierstead, Six Sigma Memory Jogger II: A Pocket Guide. GOAL/QPC: New Hampshire.
� Federico and Beaty, Rath & Strong’s Six Sigma Team Pocket Guide. McGraw-Hill: Chicago.
� Munro, Ramu, & Zrymiak, The Certified Six Sigma Green Belt Handbook (2e). ASQ Quality Press: Milwaukee.
While we provide a set of basic Microsoft Excel, Word, and PowerPoint templates for our students, online resources for LSS tool templates, instructions, and examples, are widely available. Recommended sites include:
� iSixSigma.com Six Sigma Tools & Templates, found at http://www.isixsigma.com/tools-templates/, last accessed on 01/07/2016
� MoreSteam Process Improvement & Lean Six Sigma Toolbox, found at https://www.moresteam.com/toolbox/, last accessed on 01/07/2016
For information specific to the cookie manufacturing process, we encourage students to use their own experience with baking, and supplement that experience with available online videos including:
� Tour the Most Delicious Cookie Factory in Italy!, found at https://www. youtube.com/watch?v = 8BFOab4VgRg, last accessed on 01/07/2016
� Cookie Manufacturing Machinery-mutchall.com, found at https://www. youtube.com/watch?v = q5ce2KGSqMU, last accessed on 01/07/2016
� How It’s Made Sandwich Cookies Oreos, found at https://www. youtube.com/watch?v = -i1oMwNgH2Q, last accessed on 01/07/2016
� Manufacturing Process of Cookies, found at https://www.youtube.com/watch?v = R-fFlCcIK4M, last accessed on 01/07/2016
Miller, Hill, and Miller 411
� Various other video resources obtained with the search term “Cookie Manufacturing” at www.YouTube.com, last accessed on 01/07/2016
Keith E. Miller, PhD, is an assistant professor of management in the College of Business at Clayton State University. His research interests focus on methods of process improvement, project management, and organizational change manage- ment. He has published previously as a chemical engineer, and this article is his first publication in the field of business. Keith has nearly 20 years of industry expe- rience as a military officer, management consultant, corporate master trainer, and Six Sigma practitioner, and has led or coached more than 400 process improvement projects to completion. He is a certified Project Management Professional, and a certified Six Sigma Black Belt and Master Black Belt.
Craig A. Hill, PhD, holds the Charles S. Conklin Endowed Chair and is a professor of supply chain management in the College of Business at Clayton State University. His research interests are in the area of supply chain management, in particular production planning, contract manufacturing, supply chain collaboration, and as well as innovations in education. He has published in journals such as the Journal of Operations Management, IEEE Transactions on Engineering Management and the European Journal of Operations Management. He has over 10 years of industry management and consulting experience in the area of production management, productivity improvement, quality control, and production planning.
Antoinette Miller, PhD, is a professor of psychology at Clayton State University. Her research interests range from psychophysiology to problem-based learning to case-based instruction and other pedagogical topics. She has published several of her case studies through the National Center for Case Study Teaching in Science, and has previously published articles in the Journal of Psychophysiology and Journal of Experimental Psychology: Learning Memory, and Cognition.
Copyright of Decision Sciences Journal of Innovative Education is the property of Wiley- Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.