Lean Six Sigma Project Example - Using DMAIC - Yellow Belt Training

it can be challenging when confronted with a lean Six Sigma project what is the first step that should be taken what steps should follow which tools would be the most appropriate at each phase well these are not easy questions to answer and every situation is different this video will attempt to help you create a road map that almost any project can fall it is a video informed by countless projects submitted to ssgi over many years in almost every situation following the meic framework is advised it proceeds through five phases Define measure improve analyze and control first we turn to the Define phase problem statement the Travis company leases Motor Vehicles in a 145 countries and has over 22,000 employees in 8,000 locations it is not a lowcost provider rather it attracts customers willing to pay more for exceptional service but exceptional service is not always easy to deliver consider their European call centers they receive an average of 10,000 calls a month from customers who have issues with their cars the level of service received or questions about their bills lately customer satisfaction has become a problem the voice of the customer VOC is apparently not being addressed the problem statement is this how can customer satisfaction be improved project Charter alen 6 Sigma project has been improved to address this Challenge in a pro project Charter has been written it focuses on the Project's goals scope timeline and team composition first the goal is a 20% Improvement in customer satisfaction second is scope it expresses what needs to be done and what does not for example customer satisfaction will focus only on call center issues not the price or quality of the leased vehicles third is the project timeline it will be 3 months finally an interdisciplinary team of three staff will be assigned to the project measure phase cypo diagram the second stage starts with a cypo diagram it identifies suppliers inputs process steps outputs and customers and it ensures that no variables have been ignored in the study cost of poor quality the cost of poor quality copq is instrumental in defining the impact or Consequences of any lean Six Sigma problem in this example copq can be measured by the costs incurred by the European call center when it fails to meet customer expectations when these costs are reduced or eliminated they represent benefits accured to the project benefits and costs projects need to be justified by delivering benefits in excess of their costs here we could estimate the benefits When poor quality is reduced if Travis loses 200 customers every month due to poor customer service and this leads to a loss of $1,000 in profit per Lessie per year cost of poor quality equals 200 * 12 * 1,000 or 2,400,000 in Lost profit per year we might assume that at least half of this amount could be saved by addressing customer satisfaction or $1,200,000 per year therefore this would be considered a benefit generated by the project three staff members will be assigned to the project for 3 months at a cost of $100,000 in addition it changes are estimated to cost $200,000 bringing the total cost to $300,000 when the benefit of $1,200,000 is compared to the cost of $300,000 the expected benefits far exceed the cost this can be summarized as a benefit cost ratio here it is 1,200,000 divided 300,000 or four in other words the benefits will be four times greater than the cost of the project data collection data collection is a central part of the measurement phase it begins by measuring measuring customer satisfaction here are customer satisfaction scores for Travis's European operations collected over the previous 10 months where 100 represents a perfect score the average over this 10-month period is 72.3 shortly you will discover that other data needs to be collected as we move through the project cycle process mapping in addition to process measurements it is also important to understand the flow of steps between receiving and resolving a call to accomplish this a value stream map VSM is drawn here is a simple one created for the European call center however many calls are not resolved on the first attempt when this happens it represents a SCE of inefficiency as well as a source of dissatisfaction for the customer here is the process flowchart that captures this more complex sequence of steps analyze phase root cause analysis in root cause analysis we attempt to determine the real cause of a problem for example we would like to uncover why almost 30% of the callers are dis satisfied with the service but in doing this we try to move past a simple explanation that might Focus only on symptoms as you might imagine solving a symptom is unlikely to have a meaningful or longlasting result so we search for the root cause sometimes however there may be several causes nonetheless each one needs to be uncovered seven waste where in the process can we look to find the cause or in some circumstances the cause is of the problem here is where the concept of seven waste can help the potential areas of waste include Transportation inventory motion waiting over processing over production and defects the most relevant in this situation in includes overprocess and defects overprocess occurs when an agent cannot complete a call in one session and must transfer it to another agent that agent then listens to the problem once again and collects relevant information from the customer the second category is defects they occur when invoice errors lead to additional contact with the customer and subsequent re work multivar chart one tool that is useful in uncovering information is the multivar chart process output can be measured in two ways the first is the average or mean and the second is the variation of data around this mean both are displayed in a multivari chart here is how it can help in this example the European group operates four call centers in London Paris Madrid and Milan since call Center Performance is likely to differ across locations comparing their average performance in the extent of variation across locations can prove to be helpful are they the same or do they differ and if they differ what can we learn from each here is a data for each location the second column includes the lowest customer satisfaction score measured over a period of six months the third column includes the highest value and the fourth is the average customer satisfaction score recorded over this same period here is a multivar chart illustrating the low high and average levels of performance for each of the four call centers the chart shows that London has the widest range while the range associated with Milan is quite narrow the highest average is Milan what is also apparent is that Milan not only has the smallest Range but also the highest average fishbone diagrams fishbone diagrams are another useful tool for gaining insight into a problem in decomposing it to better find its root cause or causes in this situation customer satisfaction is the presenting problem and as such it is the head of the fish from this diagram we can see all the issues that may be contributing to disappointing customer satisfaction scores [Music] five yse repeatedly asking why is another useful strategy for uncovering the layers that can lead to a better understanding of the problem in lean Six Sigma we ask five why questions each probing deeper to find the real or root cause of the problem here are five questions that can be asked in the Travis project why do you believe satisfaction Has Fallen why do you think no one has taken responsibility to address this issue why are customer service reps not adequately trained why can't we train a service rep to answer all questions why can't we use AI to help us ensure that the call is completed in one session data analysis data is important but raw data is not not enough it needs to be analyzed in a systematic way which often requires the use of Statistics so we might for example compute the average satisfaction reported previously that average is 72.3 where 100 is a perfect score data is also critical for comparing old and new ways it avoids ambiguous comparisons such as the Improvement does seem to be working or the Fulfillment process is a bit faster indeed when data is used we can be more explicit satisfaction has increased from 72 to 81 analyzing other data however the analysis phase is not limited to using statistics it also involves analyzing value hisory Maps answers to the five y or data displayed in the fishbone diagram all of these help when analyzing a problem and learning as much as possible about it next is the improve phase eliminate nonv value added steps armed with the Insight gained during the analysis phase The Next Step is Improvement nonv value added steps are eliminated to the extent possible delays errors and waste are addressed for example every effort would be made to minimize the delays in answering a customer's call even value added steps would be explored to determine if they could be made more efficient if we take a closer look at the process map for the European centers one obvious Improvement would be eliminating the need to transfer a call to a second agent another Improvement could be suggested by following up with the multivar chart exploring how Milan performs so well and then using this insight to suggest changes at the other three call centers then when reviewing the fishbone diagram it might be appropriate to consider more training when agents are better trained they may not need to transfer calls to other agents one way to address many of these Alternatives is to schedule a brainstorming session attended by relevant stakeholders during which time all of these possibilities for improvement are discussed their input can be instrumental in motivating change and limiting resistance that off often interferes with what has to be done FMEA and pokayoke during the brain storming session it might be useful to rely on FMEA and pocok failure mode and effects analysis FMEA involves staff anticipating problems before any changes are implemented and then taking steps to avoid these problems in pooke staff considers ways to prevent future errors comparing old and new Once the improvements have been made it is natural to ask if there is concrete evidence suggesting that process performance has improved in our example it could be pointed out that after changes were made customer satisfaction scores went from an average of 72.3 to 81.4 this is an improvement the goal of 20% has been reached reaching Six Sigma but have we reached a Six Sigma performance level a Six Sigma level delivers fewer than 3.4 defects per million opportunities which means that the European Center would deliver fewer than 3.4 dissatisfied customers per million customers as you can imagine achieving such results would be a significant challenge of course call center staff could be trained with this standard in mind but reaching that standard could prove very costly instead a standard less than Six Sigma would likely be more feasible so a customer satisfaction score of 81.4 might be quite reasonable control phase outputs change over time so establishing a monitoring and control system to ensure performance does not deteriorate is a critical step in lean 6 Sigma indeed any project that fails to create a monitoring and control system is one that is not complete control charts are at the center of the monitoring and control function while the statistics Behind These charts can be quite complex the basic concept is simple when constructing a chart for customer satisfaction a sample of 20 customers would be taken every month and the average would be plotted on a control chart as long as the sample mean remains within specific boundaries the process would be considered to be in control to construct this control chart we would use the long-term average output as the Target or in this situation we could use the average over the last 6 months once the changes have been implemented the upper boundary is called the upper control limit and the lower boundary is the lower control limit here is the control chart for this project notice that the sample means in this sample are within the upper and lower control limits so customer satisfaction scores can be considered under control that's it this is a very brief introduction to the way we would organize and manage a lan6 sigma project while your project may involve other tools and is likely to go into greater statistical depth the basic format and Direction described here can guide almost any lean Six Sigma project to recap the basic issues of a lean6 sigma study include follow a framework clearly Define the problem measure problem parameters by collecting data analyze the data with such tools as multivar charts and value stream Maps improve the problem and finally control the process to ensure that it continues to meet goals for a full range of courses from yellow to master black belt as well as courses that focus on operational excellence scrum management agile and Leadership visit us at Six Sigma Global institute.com thanks for watching