Sales Department - nargund.com

Sales Department - nargund.com

About This Project This project is a simulation of actual occurrences Covers key six sigma concepts including Applying the DMAIC approach to process improvement Identification and selection of process improvement opportunities Utilizing Statistical Analysis and Tests Addressing/Improving Customer Satisfaction Cost Savings & Ongoing Financial Benefits Seeks to accomplish key outlined objectives Provide Detailed Explanations Throughout Illustrative Analysis Comprehensive Use of Recommended Tools Effective Resolution/ Final State Presenters Knowledge of Six Sigma Methodology DSL EASTERN DIVISION IN T t N E uc io n P ro je c t M d T e R R A P E on D ti N la O I al T A st L L In A ST s t G h o Prepared By: Joe Banks & Cecilyn Cayetano Definition: Ghost Installation (GIs): Installation attempt in which the installer found no one available on-site once he/she arrived to perform an installation; resulting in a defective installation job. DMAIC Project Description Resource Requirements Project No: Project Name: Green Belt or Black Belt Master Black Belt Finance Partner: Champion: General Information 216 Ghost Installation (GI's) Reduction Project Black Belt NA DSL-East Unit Robert Price, DSL EVP Location: Atlanta,GA Business DSL-Eastern Division Segment: Installation Department Business Primary: To reduce the amount of Ghost Installations Objective: by 33% within 6mo. Secondary: To increase rate of completed jobs by 5% or more above current levels. Customer Internal: Completed Installations, Daily Install CTQ(s): Routes Completed, Low-Reinstallation Rates Team Members: Name

Function % Time External: On-Time Installs, Communication, Correct Installation, Pleasant Expirience Initials Cecilyn Cayetano Project Analyst 100 CC Stakeholders Joe Banks Project Analyst 100 JB Current Process Supporting Members: Name Installers David Morrow (Unit Supervisor) Jennifer Bell Resources Consultants Data & Stats Financial Analysis Installers, DSL-East, DSL Division Installation Z score = 1.0803, GI's DPMO= 150,000, As of Date: N/A N/A N/A N/A N/A N/A 6/16/2010 Expected Benefits: Higher Customer Satisfaction, More Customers Served Financial Benefits $: $99,700 (Revenues and Cost Savings) Organization Benefits: Increased Billing Revenues, Less Employee Rework Problem: Objective: Scope: Measurements: During a review of year over year comparisons of DSL-East installation reports it was discovered that the GI rate across the DSL-Eastern Divisions territory is trending an all time high of 15%, causing repeat installs and lost customers. To reduce the rate of GIs (Big Y) below the upper specification limit of 10%, which will in turn increase the rate of completed jobs back to normal levels of 90% or more. Metrics (unit of measure): The rate of successfully completed Installations, nondefective. Completed installations >90% (5% improvement). In conjunction with the rise in GIs there has also been a 10% increase in customer complaints due to the missed installation appointments. It is our goal to reduce the rate of Ghost Installations from 15% of total installs to below 10%, a 33% reduction resulting in DPMO < 100,000 and a yield of 90%. Defect Definition: Installation attempt in which the installer found no one to be available on-site once he/she arrived to perform an installation resulting in a defective installation job. Installations achieve a long term process sigma of > 2.7. Eliminate the 10% increase in customer complaints. DMAIC DSL-East Total Install Process Key Point: Our Projects Focus will be in the DSL Installation Department Id like to order DSL Service

Series 1 Series 2 Overview of DSL Series of Events Sales person receives call with new install order RLD confirms equipment inventory, installers schedule, availability of date & time Project Selection: Sales forwards the new order to RLD for 48 hrs confirmation RLD receives order and begins processing RLD Sends YES or No confirmation back to sales within 48hrs. RLD sends order to equipment warehouse for processing Several departments within the unit have improvement areas and possible projects. We selected this project by using a Project Prioritization Matrix. Prioritization Scores: scores are weighted Series 3 Equipment warehouse receives order and begins physical processing Equipment Warehouse sends YES or No instock confirmation back to RLD within 72hrs If equipment is in stock the warehouse packages it and moves it to the Installation dock 72hrs prior to install Our Focus Series 4 Installation dock places equipment in proper bay and organizes by date of delivery order When install date arrives it is placed on the proper truck Unit Project Score Sales A 3.1 Sales B 3.6 Warehouse C 2.9 Warehouse D 2.3 Installation E 4.1 Installer takes truck and goes to perform insatall * RLD = Regional Logistics Department DSL-Easts GI Defects vs. Other Unit Key Point: The DSL Easts GIs are Higher than Normal DMAIC Project Validation: From the historical data we can see that the amount of DSL-East GIs is at an all time high. The DSL-West Division is performing normally. Scale zoomed in for impact. DSL-East GI Trend 30% 25% 20% 8% Historical Baseline 15% Installs GI's 10% 5% 0% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 DSL-West GI Trend Scale zoomed in for impact. 30%

25% Our Focus 20% 6.5% Historical 15% Baseline 10% 5% 0% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Installs GI's Recent Changes for DSL-East Specifically Key Point: Geographical Expansion Has Expanded DMAIC Service Areas for Cities Serviced by DSL-East Units Fact: Eastern U.S. Cities Experience Explosive Population Growth During the Recent Housing Boom: Jacksonville, FL 10.6% Orlando, FL 21.5% Charlotte, NC 24.9% Nashville, TN 11.0%, Atlanta, GA 29.2% Miami, FL19.5% Raleigh, NC 40.7% With the recent housing expansion in the United States we have seen new neighborhoods and rural expansion surrounding many previously smaller eastern US cities. This is in contrast to the West having greater population than geographical growth in major *Source: USA Today cities with less rural territory expansion, this evidenced by http://www.usatoday.com/news/nation/census/2010-06-22-census_N.htm Scope, VOC & VOB So youre going to be 30mins late Suppliers 1 Sales Agents 2 Internet Order System 3 Dispatch Input SIPOC Process (High Level) Manually Relays Installion Orders and Changes There are several key factors that accurate, timely, and courteous installations affect. All of which add to the success of the business, the business wants... High Customer Satisfaction Potential Referrals W.O.M To Secure New Billings Fewer Re-Installs (Rework) Reduce Equipment Restocks Reduce Customer Complaints Enter Location into GPS Confirm Travel Distance and Time Estimates Choose Traffic Route Check for Customer Communication Details Call Customer (if requested by customer) Confirm Location with Customer

Confirm Approx. Travel Time with Customer Travel to Customer Site Attempt Installation End Point: Enter Data Into Completion System and email. DMAIC I had my phone with me The jerk never called!!! Output Customers Start Point: Completed Retrieve Installation Orders via Installation Orders New DSL Customers Installation Daily Installation Order System (DIOS) Operation or Activity New Customer Installation Orders Data Collection Hub Review Order Details Sales Data Voice of the Business: Key Point: Customers are Complaining; Theres a Problem Signed Work Order Customer/Business Partners/Others Department Email Detailed Installation Data Regional Logistics Department Voice of the Customer: We used the call center database to retrieve details on missed installations. The data contains customer comments about why the install was missed, the order info that was provided to the installer originally, as well as the installers reference code for the Ghost Installation. I d o n t c a re i f yo u ' re st u c k i n t raffi c . I h ave to l e ave i n 3 0 m i n s ! ! ! KPIV,KPOV, & Data Collection DMAIC Key Point: KPIVs: Traffic & Distance, KPOV: Completed Jobs Cause & Effects Matrix Weather Reaching Reaching Customers Customers Traffic From the results of our cause and effects matrix we can see that the key inputs (xs) to the process are estimating traffic delays and effectively measuring the distance from location to location ahead of leaving for the installation. 0 = no possible effect, 3 = possible effect, 6 = known moderate effect, 9 = known large effect Other Location Distance Distance Top 3 Arrival Requirements Find Location On Time Familiar w/Area 10 8 6 Scores Available Customer

1,3,9 Importance Process Steps Process Input Correlation of Input to Output Total Scores Call Customer Operator Est. Traffic Times Operator Measure Distance GPS Causes for Ghost Installations Based on the coded data retrieved from the data entry system it appears that the most common cause for missed appointments as stated by installers is traffic (construction, detours, accidents), followed by distance (location to location distance), communication (cannot reach customer), etc... 200 Process Outputs 9 6 6 3 9 6 0 9 6 C&E Matrix Results 186 180 144 160 140 114 120 100 80 60 40 20 0 Est. Traffic Times Measure Distance Call Customer 114 186 144 Measure System Analysis DMAIC Key Point: The Overall Process is Normally Distributed The frequency histograms below helped us determine that our data is normal. On the left we can see that the combined % of completed installations across both divisions is normally distributed at a rate of about 88%. To the right is the completion rate for both divisions shown independently; DSL-Easts mean is below the LL specification of 90%. 3 2 1 0 Normal Mean 0.8852 StDev 0.02686 N 30 Mean 0.8852 StDev 0.02686 N 30 Normal 4 3 2 1 0 9 Frequency Frequency Frequency Frequency

4 East vs. West Completed Installs Normal Histogram of Total Installs 5 5 Histogram of Total I nstalls 0.84 0.84 0.86 0.88 0.90 0.92 % of Total Installs Completed 0.86 0.88 0.90 0.92 % of Total Installs Completed 0.94 0.94 8 9 7 8 6 7 5 6 4 5 3 4 2 3 1 2 0 1 0 East vs. West Completed Installs Variable Normal East % of Completed Installs West % of Completed I nstalls Variable Mean StDev N East % of Completed Installs 0.8548 0.09309 31 West % of Completed I nstalls 0.9135 0.06892 31 Mean StDev N 0.8548 0.09309 31 0.9135 0.06892 31 Wests Benchmark 0.7 0.8 0.9 1.0 % of Total Completed Installs 0.7 0.8 0.9 1.0 % of Total Completed Installs MSA Continued Key Point: Xs & Ys are in Control, Yet Not Meeting Process Specs DMAIC P Chart of Defectives 0.30 Control Charts Analysis UCL=0.2988 The P Chart corresponds with the histograms that about 15% of the installations are actually defective. 0.25 Proportion 0.20 _ P=0.1481 0.15 0.10 The sample data used for the I-MR charts of traffic and distance (KPIVs) shows us that the data is in control, although we know by the rate of defective installations (15%) that the process isnt meeting specifications (<10%). Defectives Baseline 0.05 0.00

LCL=0 1 6 11 16 21 26 31 Sample 36 41 46 51 I -MR Chart of DSL-East Traffi c Data I-MR Chart of DSL-East Distance Before_ UCL=46.35 UCL=93.7 50 _ X=35.3 0 I ndividual Value I ndividual Value 100 11 16 21 26 Observation 31 36 41 Moving Range 80 Baseline for logged traffic times UCL=71.75 60 40 __ MR=21.96 20 0 LCL=0 1 6 11 16 21 26 Observation 31 36 41 46 _ X=21.11 20 10 46 LCL=-4.13 1 6 11 16 21 26 Observation

31 36 41 30 Moving Range 6 30 0 LCL=-23.1 1 40 Baseline for logged distance traveled UCL=31.01 46 20 __ MR=9.49 10 0 LCL=0 1 6 11 16 21 26 Observation 31 36 41 46 MSA Continued Key Point: GPSs are Performing their Desired Function; Installer Can DMAIC Trust the Route Information Given to Them by the GPS System Testing The System: We evaluated the measurement system (GPSs) used to determine the distance from the dispatch location to a fueling station with a known distance of 2mi. Weve imposed a tolerance level of .1 mi, and performed 50 observations. GPS Distance Repeatability Gage name: Date of study: Reported by: Tolerance: 0.1 Misc: Distance The Result: Accept Ho Run Chart of Distance Distance 2.02 2.01 Ref + 0.10 * Tol 2.00 Ref 1.99 Ref - 0.10 * Tol 1.98 1 6 11 Basic Statistics Reference 2 Mean 1.999 StDev 0.0120 6 * StDev (SV) 0.0717 Tolerance (Tol) 0.1

16 21 26 Observation Bias Bias T PValue (Test Bias = 0) 31 -0.001 0.7159 0.477 36 41 46 Capability Cg 0.28 Cgk 0.25 % Var(Repeatability) 71.74% % Var(Repeatability and Bias) 81.62% The P Value in the measurement system is .477 suggesting that no bias is present in the measurement system. This result preserves the H0; there is no difference in the results the GPS provides over multiple uses /users. Also, we noticed that many of the observations plotted on the run chart appear evenly distributed both above and below the reference of 2mi. The difference of the largest and smallest values = .04 which is less than our tolerance level of .1 signaling the gage (GPS) and its user(s) may be considered accurate and repeatable and therefore shouldnt be improved. This conclusion leaves us with the unanswered question of why is distance the #2 reason for GIs? MSA Continued Key Point: GPSs are Performing their Desired Function, Estimating the Area Traffic Isnt Proving to be a Consistent Method Across Installers Testing The Operators vs. The System: Gage R&R (ANOVA) for Traffi c Time Est. Components of Variation Traffi c Time Est. by Loc 100 % Contribution Percent % Study Var 50 50 25 Understanding The Results: 0 Gage R&R Repeat Reprod 1 Part-to-Part Sample Range R Chart by Operator 1 2 _ R=18.67 0 LCL=0 3 1 2 Loc 3 1

2 3 50 25 0 1 Xbar Chart by Operator 2 3 UCL=41.39 20 __ X=22.30 0 LCL=3.20 1 2 3 1 2 Loc 3 1 2 Operator 3 Loc * Operator Interaction 40 2 3 45 Average 1 In the Components of Variation graph (located in the upper left corner), the percent contribution from Total Gage R&R (97.97) is larger than that of Part-To-Part (2.03). Thus, most of the variation arises from the measuring system (estimating traffic times) not the locations themselves. 3 UCL=48.05 20 2 2 Loc Traffi c Time Est. by Operator 3 40 1 Sample Mean Three locations were selected that represent the expected range of the process variation. Three operators measured the expected traffic times for the three locations (assuming no special circumstances), three different days per location, in a random order. Reported by: Tolerance: Misc: Gage name: Date of study: 0 DMAIC Operator 1 2 30 3 15 1 2 Loc

3 In the Xbar Chart by Operator most of the points in the X and R chart are inside the control limits, indicating the observed variation is mainly due to the measurement system. In the By Part graph (located in upper right corner), there is little difference between parts, as shown by the nearly level line. The Total Gage R&R accounts for 98.98% of the study variation. The measurement system of individual drivers estimating traffic times/conditions is unacceptable and should be improved. MSA Continued Key Point: The Process is 5% below the Lower Specs, We DMAIC Now Have Clues as to Why Process Capability of DSL-East Completed Installs LSL USL The Result: Within Overall Process Data LSL 0.9 Target * USL 1 Sample Mean 0.859 Sample N 30 StDev(Within) 0.0255495 StDev(Overall) 0.0224914 Potential (Within) Capability Z.Bench -1.60 Z.LSL -1.60 Z.USL 5.52 Cpk -0.53 O verall Capability Z.Bench Z.LSL Z.USL Ppk Cpm 0.81 0.84 0.87 0.90 0.93 0.96 0.99 Observed Performance PPM < LSL 933333.33 PPM > USL 0.00 PPM Total 933333.33 Exp. Within Performance PPM < LSL 945723.14 PPM > USL 0.02 PPM Total 945723.16 Exp. Overall Performance PPM < LSL 965842.31 PPM > USL 0.00 PPM Total 965842.31 -1.82 -1.82 6.27 -0.61 * Here weve displayed the Current State of DSL-East Completed Installs, we can see that the DSL-East division is currently completing only 85% of their installations on average, we can expect performance below our specified (LSL) completion rate of 90%, 96% of the time. This process is incapable of meeting the specs and must be corrected! Process Capability Key Point: The Process is Incapable of Meeting Specification DMAIC Capability Analysis: Running a capability analysis we confirmed that the DSL-East division is yielding 85% of their installations on average, with 15% of all installations being defective, producing 150,000 defectives per million opportunities. With a yield of less than 6%, and a dismal long term process sigma of .1, we must reduce process variability and move into the spec range. Six Sigma Capability Analysis (Before) Defective Number of units Number defective observed Proportion defective % Yield 20,000 3,000.0

0.1500000 85.00% Capability Measure Short Term DPU DPPM Yield Cpk Ppk Process Sigma 0.15 150,000.0 85.00% 0.35 1.04 Variables, (Normal) Average 0.86 Stdev. 0.0255 USL 1.00 LSL 0.90 Cp 0.652 Cpk (0.535) 6.65% 6.65% We currently run the risk of being out of the spec range 95% o f t h e ti m e . Capability Measure Long Term (+1.5) DPU DPPM Yield Cpk Ppk Process Sigma 0.9457232 945,723.2 5.43% (0.53) (0.1) * Capability Analysis Courtesy of Thomas A. Little Consulting * Defects = Defectives : There are no defects for GIs, the job is simply defective if the installer found no one present or no location to perform the install . Current State Process Map DMAIC Key Point: Missed Installs Causes Rework & Increased Costs Process Map Organization Operation Sequence or Time Start or End Organization 1 Confirm Travel Distance/With Time Est. Choose Traffic Route Check for Customer Communication Details No Decision Travel to Customer Site Data Stored VOC Opportunity Customer Available ? Attempt Installation Yes Starts the Process VOC Opportunity Cost of Poor Quality

Costly Rework Worker Inefficiency (overstaffing) Equipment Restocking Lost Business Opportunities Low Customer Satisfaction Rise in Customer Complaints Low Return on Investments Confirm Approx. Travel Time w ith Customer No Wait for customer or Wait for Next Job Travel to Customer Site Log the GI into Sytem Causes Rework Data Ends the process Causes Rework Attempt Installation Causes Rework Ends the process Customer Available Yes Perfom Installation No Log the GI into Sytem Ends the process Ends the process Data Confirm Location w ith Customer Perfom Installation No Data Poor execution of this process leads to Is Customer Ready for Install Yes Data Review Order Details Wait or Delay Data Retrieve Installation Orders via Daily Installation Order System (DIOS) Operation Data VOC Opportunity Ishikawa (Fishbone) Diagrams Why is traffic DMAIC Key Point: Brainstorming on Possible Causes of KPIVs Why is traffic causing Ghost Installations? Environment

Construction Machinery Poor Weather Materials Source for Traffic Info GPS has no traffic function Road Blocks Fuel Needs Missing Equipment No Standard Policies No way to gauge changes Waiting in Traffic Unfamiliar with Area Radius too wide Route Errors Measurement Environment Machinery Slow Responses from GPS Too Much Traffic Severe Weather Determining Distance GPS Use Not Mandatory No Local Familiarity Training Use Personal Experience Measurement Address not visible GPS Cannot Find Loc. Wide Service Area Poor Time Estimates Methods Materials Doesnt Display Traffic Methods Training Installation Times (am,pm) Man Man Are the GPS Systems Out of Date? FMEA Key Point: Critical Effects: 1) Est. Traffic 2) Est. Distance 3) Customer Communication DMAIC Failure Modes & Effects Analysis: Walking through the FMEA process has allowed us to assign values to critical process inputs so that we can prioritize our corrective efforts. FMEA Objective, scope and goal(s): To identify critical improvement needs and to understand the improvement implementation risks A Pr oce s s Ste ps / Input Pote ntial Failure M ode(s ) Pote ntial Effe ct(s ) of Failur e What is the process step and input under investigation? In w hat w ays does the Key Input go w rong? What is the impact on the Key Output Variable (Customer Requirements) ? Estim ating Traffic Conditions Under estim ates traffic conditions Ins taller arrives

late and m is s es appointm ent Estim ating Traffic Conditions Over estim ates traffic conditions Estim ating Tim e Under estim ates tim e to arrive at location Ins taller arrives too early and m us t wait to begin work Ins taller arrives late and m is s es appointm ent Estim ating Tim e Over estim ates tim e to arrive at location Estim ating Dis tance Under estim ates dis tance to arrive at location Estim ating Dis tance Over estim ates dis tance to arrive at location Cus tom er Com m unication Requirem ents Ins taller cannot reach cus tom er Ins taller arrives too early and m us t wait to begin work Ins taller arrives late and m is s es appointm ent Ins taller arrives too early and m us t wait to begin work Ins taller cannot get needed info S E V Pote ntial Caus e (s )/ Me chanis m (s ) of Failur e What causes the Key Input to go w rong? 8 No SOP for getting updated traffic info 2 No SOP for getting updated traffic info 8 Equipm ent arrival tim es are not accurate 2 Equipm ent arrival tim es are not accurate 10 Operators choose alternate routes 2 Operators choose alternate routes 8 Cus tom er contact is by reques t only Curre nt De s ign/Pr oce s s P Contr ols R O What are the existing B controls and procedures (inspection and test) that prevent the cause of the Failure Mode? D E T R P

N Re com m e nde d Action(s ) What are the actions f or reducing the occurance of the cause or improving detection? Re s pons ibility Who is responcible f or implementing reccommended actions? 10 None 10 800 Update equipm ent to provide real tim e traffic updates Jennifer 6 None 6 72 Update equipm ent to provide real tim e traffic updates Jennifer Ah Ha Installers Discretion Ha384 Installers 6 Us e travel tim es givenAh 8 Tes t Discretion m ultiple m fg's Causes Errors Distance by GPS forin the m os t Measurements!!! accurate equipm ent 6 Us e travel tim es given by GPS 4 8 Ins tallers descretion as to us e GPS Route, no SOP 8 640 Create SOP to us e GPS routes /directions David 6 Ins tallers descretion as to us e GPS Route, no SOP 8 96 Create SOP to us e GPS routes /directions David 6 Cus tom ers can 10 reques t or decline to be contacted by ins taller prior to arrival 48 Tes t m ultiple m fg's for the m os t accurate equipm ent Joe & Cecilyn Joe & Cecilyn 480 Create policy that Jack cus tom ers m ust be contacted before ins taller proceeds to their location Root Cause & DOE Analysis Key Point: Traffic & Distance Have the Most Significant Effect on DMAIC Travel Times; Also GIs vs. Customer Complaints p-value = .000 Interaction InteractionPlot Plotfor forTime Time Data DataMeans Means 2009 Common Causes of GIs

Defect Inputs: Pareto 1206 0.3 The Pareto chart illustrates that over 80% of GIs are due to the top 3 causes (xs). 798 0.2 498 312 0.1 0 Traffic Weather Distance Other 16 16 32 32 Distance Distance ABC AB AC BC 4 6 8 Standardized Effect 10 12 This chart indicates that all the main effects are significant although weather (temp.) much less than the others. We can also see the interactions that are significant are Traffic and Distance or all 3. Versus Fits In the analysis of variance table Traffic * Distance (p = 0.021), and main effects are significant. 100 90 Percent Name Traffic Distance Temperature 50 10 50 0 -50 1 -100 -50 0 Residual 50 100 0 60 Histogram 120 180 Fitted Value 240 100 7.5 50 5.0 2.5 0 -50 -80 -40 0 40 Residual The p-value in the Analysis of

Variance table (0.000), indicates that the relationship between defects (x) and customer complaints (y) is statistically significant at an alpha level of .05. Versus Order 10.0 0.0 The non parallel lines found across all the interactions indicate that at high levels of any 2 of the factors (traffic, distance, temp.) the response (travel time) will increase. Regression: Reject H0 Complaints vs. Defects Frequency C Distance Distance 11 22 33 Interaction Plot: Time Scale: 3= High, 2= Med, 1= Low Normal Probability Plot 99 B Term 24 24 Traffi cc Traffi 11 22 33 16 16 2.36 A 33 24 24 DOE: Pareto Effects Factor A B C 22 Traffi cc Traffi Comm. Pareto Chart of Effects 2 11 Temperature Temperature (response is Time, Alpha = 0.05) 0 33 32 32 Traffic - 40.2% Dist. - 26.6% Comm. - 16.6% 186 22 Residual 0.4 11 Residual 0.5 80 1 5

10 15 20 25 Observation Order 30 35 Because there is significance in the rate of complaints versus GIs we must reject H0: That there is no significance between the two occurrences, and accept the alternative. Future State Brainstorming Key Point: 3 Main Areas Identified for Improvement Opportunities DMAIC Potential Solutions * solutions in green text can be implemented immediately Technology: Training: GPS Features Route Selection Time Management Quarterly Service Area Briefings Construction Starts Downtown October 2nd Policy & Procedures: No Discretionary Routes Require Customer Confirmation Before Traveling to Site Post New Obstructions Assign Drivers as Locally as Possible to Their Neighborhood Hand-Off Routing (Flexible AdHOC Dispatching) Upgrade to GPS w/ Live Traffic Conditions Upgrade to GPS that provides alternative routing Text Weather Alerts Automated Calling Confirmation System Future State Brainstorming When possible Why not move up waiting customers by dispatching close-by waiting installers? DMAIC Key Point: Top Three Solutions Identified at Kaizen Event Prioritization of Solutions Creates Higher Customer Satisfaction 0.1 5 8 8 8 10 8 0.4 9 8 6 8 5 2 0.3 7 3 9 5 1 1 Overall Rank Most Effect on Defect Reductions 0.2 3 8 7 4 10 7 Final Weighted Score Cost Less than 60K Solutions Automated Calling Confirmation

New GPS w/ Live Traffic Routing Hand-OffDispatching Require Cust. Confirmations Assign Local Drivers & Routes Time Management Training Quickly Implemented Weighted Benefits 6.8 6.5 7.3 6.3 5.3 3.3 2 3 1 4 5 6 Pilot Testing of Solutions Key Point: Over Our 3 Month Trial Period Major Overall Gains Have Been Made Regarding Installation Attempts 100 0 Cum % Distance Communication 798 498 26.6 16.6 40.2 66.8 83.4 800 50 Weather 312 10.4 Other 186 6.2 93.8 100.0 Cum % 0 Distance 502.5 35.3 Traffi c 399.0 28.0 35.3 63.3 77.9 91.3 Other 124.0 8.7 UCL=93.7 50 _ X=35.3 0 40 16 21 26 31 Observation 0 36 41 46 1 6 11

_ X=21.11 20 0 LCL=-4.13 16 21 26 31 Observation 36 41 21 26 31 Observation 36 41 46 Flexible Dispatching Reduced Ave. Distance UCL=39.86 Traveled 46 40 I ndividual Value I ndividual Value UCL=46.35 11 16 _ X=18.15 20 0 LCL=-3.55 1 6 11 16 21 26 31 Observation 36 41 46 Ave. In-Traffic Times Reduced by -24% -8mins Logged Traffic Delay Counts Down by 67% Distance Logged: I Chart of DSL-East Distance After 40 6 LCL=-17.6 I Chart of DSL-East Distance Before_ 1 Traffic Times: _ X=26.8 LCL=-23.1 11 52% Overall Defective Reduction Achieved Traffic, Communication, and Distance Have the Most Improvement 100.0 Updated GPS Technology w/ Alternate Traffic Views UCL=71.2 Reduced Traffic Counts

80 I ndividual Value I ndividual Value Weather Communication 208.0 190.9 14.6 13.4 I Chart of Traffi c After 100 6 0 Defect_1 Count_1 Percent I Chart of Traffi c Before 1 Percent 50 Defects Reduction: 1600 Count_ 1 1500 Traffic 1206 40.2 Forecasted Reductions Shows Well Beat Our Historical Benchmark!!! GI Defects After 100 Percent Count GI Defects Before 3000 0 Defect Count Percent DMAIC Ave. Distance Reduced by -14% -3mi Logged Traffic Delay Counts Down by 37% Hypothesis Testing Key Point: The Percentage of Completed Installs Has Risen Into the Specification Area. Boxplot of Before Completed I nstalls, After Completed I nstalls 1.00 Percent Complete 0.95 In-Spec Area 0.90 Non-Spec Area 0.85 0.80 Before Completed Installs After Completed Installs Value Plot of Average Completed Installs 1.00 Percent Complete 0.95 After Improvements 0.90 Before Improvements 0.85 0.80 Before Completed Installs

After Completed Installs Two Sample T-Test The Boxplot and Value Plot of before and after completed installations shows the expected average % of completed jobs has risen to meet specs of > 90% , and will slightly surpass the prior historical baseline of 92%. DMAIC Improved Yield Analysis DMAIC Key Point: Process Capability is Higher and Complaints Will Decline by > 30% and Below All Historical Levels by Year End Six Sigma Capability Analysis (Before) Defective Number of units Number defective observed Proportion defective % Yield 20,000 1,424.0 0.0712000 92.88% Capability Measure Short Term DPU DPPM Yield Cpk P pk Process Sigma I m te l l i n g yo u H o m e r, t h e g u y w a s o n ti m e a n d d i d a g re at j o b ! 0.0712 71,200.0 92.88% 0.49 1.47 Post Improvement Capability Variables, (Normal) Average 0.94 Stdev. 0.0278 USL 1.00 LSL 0.90 Cp 0.600 Cpk 0.457 6.65% 6.65% With the defective installs slashed by 52% we can expect to achieve an acceptable yield of 93% of all jobs completed without GI issues. Capability Measure Long Term (+1.5) DPU DPPM Yield Cpk P pk Process Sigma 0.0853410 85,341.0 91.47% 0.46 2.9 We can also see below that customers complaints are decreasing in response to the improvements in service delivery. Year Over Year GI Customer Complaints 400 350 300 250 200 150 100 50 0 Jan May 2010 Improvements Implemented! Feb Mar Apr Customer Complaints Before 2009

* Capability Analysis Courtesy of Thomas A. Little Consulting May Jun Jul Customer Complaints During 2009 Aug Sep Oct Nov Dec Customer Complaints After 2010, 2011 Est. Forecasted Process Capability DMAIC Key Point: Capability Analysis Shows on Average We Can Now Expect to Meet Service Specifications Process Capability After This Process Simply Missed the Mark Before Our Analysis. LSL USL Within Overall Process Data LSL 0.9 Target * USL 1 Sample Mean 0.93803 Sample N 30 StDev(Within) 0.0277588 StDev(O verall) 0.0250994 Potential (Within) Capability Z.Bench 1.29 Z.LSL 1.37 Z.USL 2.23 Cpk 0.46 Overall Capability Z.Bench Z.LSL Z.USL Ppk Cpm 0.90 Observed Performance PPM < LSL 33333.33 PPM > USL 0.00 PPM Total 33333.33 Exp. Within Performance PPM < LSL 85340.59 PPM > USL 12792.65 PPM Total 98133.24 0.93 Process Capability Before LSL USL Within Overall Process Data LSL 0.9 Target * USL 1 Sample Mean 0.859 Sample N 30 StDev(Within) 0.0255495 StDev(Overall) 0.0224914 Potential (Within) Capability Z.Bench -1.60 Z.LSL -1.60 Z.USL 5.52 Cpk -0.53 Overall Capability Z.Bench Z.LSL

Z.USL Ppk Cpm 0.81 0.84 Observed Performance PPM < LSL 933333.33 PPM > USL 0.00 PPM Total 933333.33 0.87 Exp. Within Performance PPM < LSL 945723.14 PPM > USL 0.02 PPM Total 945723.16 0.90 0.93 0.96 Exp. Overall Performance PPM < LSL 965842.31 PPM > USL 0.00 PPM Total 965842.31 0.99 0.96 0.99 Exp. Overall Performance PPM < LSL 64863.79 PPM > USL 6774.83 PPM Total 71638.62 -1.82 -1.82 6.27 -0.61 * Now We are on Target and Ready to Fully Implement the Solution! 1.46 1.52 2.47 0.51 * Updated Process Map w/SOPs DMAIC Key Point: Key Process Improvements: 1) Increased Efficiency 2) Increased Customer Contact 3) Key SOPs are Now in Place Updated Process Map Organization Operation Sequence or Time Start or End Operation Wait or Delay Decision Data Stored SOP-Process Improvement Steps Retrieve Installation Orders via Daily Installation Order System (DIOS) Review Order Details Confirm Travel Distance/With Time Est. (GPS) Log the GI Into Sytem Confirm Customer Communication Details Attempt to Reschedule Customer Call Customer No Is Customer Ready for Install? SOP-Customer Interaction

Yes Confirm Location w ith Customer Confirm GPS Travel Time w ith Customer Travel to Customer Site Gather VOC No Log the GI Into Sytem Gather VOC Yes Customer Available ? Perform Install Gather VOC Data Data Get Closest Job From Dispatch Attempt to Reschedule Customer Data Data Get Closest Job From Dispatch Traffic Route via (GPS) Attempt Installation Process Monitoring via Control Charts DMAIC Key Point: Defectives and Customer Complaints Due to GIs are Now In Control After the Improvements P Chart of Defectives Before P Chart of Defectives After 0.3 UCL=0.2988 UCL=0.1682 0.16 0.12 _ P=0.1481 0.1 Proportion Proportion 0.2 0.08 _ P=0.0642 0.04 0.0 LCL=0 1 6 11 16 21 26 31 Sample 36 41 46 0.00 51

LCL=0 1 6 % of Customer Complaints Due to GI's Before 0.5 11 16 21 26 31 Sample 36 41 46 51 % of Customer Complaints Due to GI's After 0.4 UCL=0.4920 UCL=0.3564 0.3 _ P=0.298 0.3 0.2 Proportion Proportion 0.4 _ P=0.19 0.2 0.1 LCL=0.1040 0.1 1 6 11 16 21 26 Sample 31 36 41 46 LCL=0.0236 0.0 1 6 11 16 21 26 Sample 31 36 41 46 Financial Benefits Summary Key Point: Through Project Improvement Efforts We Have DMAIC Created $276,000 in Total Benefits for the next 12mo., and $196,400 in Reoccurring Annual Revenues Annual Revenues (based on yr/yr comp. rates) 2011 Savings Installs Labor New Billings Completions New Billings from Reffe red Customers

167,000 29,400 196,400 Wages Overtime $ 196,400 36,000 9,000 45,000 $ 45,000 19,800 $ 19,800 3000 $ 3,000 $ 67,800 Sales (one time) Equipment Sales due to Installs 11,800 Total Revenues $ 11,800 $ 208,200 Project Cost (one time non-reoccurring) Raw Mat (subject to mrkt prices) Fuel Warehousing Labor Indirect Labor Direct Labor Overtime Overtime Reduction Total Savings 3300 21,000 6,500 30,800 $ 30,800 Equipment GPS Upgrade Automated Calling System Misc 32,000 12,500 3,700 48,200 $ 48,200 Training Financial Summary 12-Mo. Net Project Gains $ 127,089 Annual Revenue Gains $ 196,400 Quarterly Revenue Gains $ 49,100 Monthly Revenue Gains $ 16,367 Total 2011 Benefits Travel Print Materials Food Misc. 1200 480 109 230 92 2,111 Total Cost $ 2,111 $ (81,111) (Ex-Equipment) $ 276,000 QUESTIONS or COMMENTS? DMAIC Thank You for Your Time Ghost Installation Reduction

Recently Viewed Presentations

  • I am a Dalek - Our Happy School

    I am a Dalek - Our Happy School

    I AM A DALEK By: Nathan Buttigieg Class: Year 5 Roses Hello, my name is Nathan and today I'm going to tell you about a book called I am a Dalek. PLOT Rose and the Doctor want to go on...
  • The Early Evolution of Protostars Neal J. Evans

    The Early Evolution of Protostars Neal J. Evans

    The Early Evolution of Protostars Neal J. Evans II with much help from Mike Dunham, Lori Allen, Melissa Enoch, Jeong-Eun Lee, Joel Green, Hyo-Jeong Kim
  • Charging - Ms. Rousseau&#x27;s Classroom

    Charging - Ms. Rousseau's Classroom

    Millikan hypothesized that by measuring the total charge on any oil drop, it . would be an integral multiple of the elementary charge (there was no way of knowing the excess/deficit of e- on a droplet). He used an electrical...
  • www.usscouts.org

    www.usscouts.org

    Venturing Leader Specific Training Reminder to instructors: Check the notes pages of this presentation for the complete text of the Venturing Leader Specific Training This is a "hidden slide" and will not show in the presentation.
  • Personal Protective Equipment - EHS DB.com

    Personal Protective Equipment - EHS DB.com

    PERSONAL PROTECTION EQUIPMENT. COURSE OBJECTIVES. Have a basic understanding of: The purpose and basic concepts of PPE. When PPE is necessary. Different kinds of PPE available. How to wear and maintain PPE. Limitations of PPE. How PPE can protect from...
  • Investigating disease - Vivienne Clark

    Investigating disease - Vivienne Clark

    Investigating disease . Lesson objective - to learn what well-being and disease means. Different definitions of health . Blaxter (1990) examined how people defined health. She carried out a survey of over 9000 people. Relating health to PIES .
  •  Grup int - elevi de liceu  Discipline implicate

    Grup int - elevi de liceu Discipline implicate

    Grup ţintă - elevi de liceu Discipline implicate - matematică, informatică, chimie, biologie, ecologie OBIECTIVE: implicarea activă a elevilor în viaţa comunităţii; promovarea interdisciplinarităţii; realizarea unui studiu predictiv cu privire la cantitatea de PET-uri ce se adună într-un an şcolar...
  • Magmatism of the Snake River Plain - Yellowstone region ...

    Magmatism of the Snake River Plain - Yellowstone region ...

    Snake River Plain - Yellowstone region: Implications for continental lithosphere evolution above a mantle plume ... evolution Development and extrapolation of understanding of large igneous systems Magmatism of the Snake River Plain - Yellowstone region: Implications for continental ...