iSixSigma

Customer service errors

Six Sigma – iSixSigma Forums Old Forums General Customer service errors

Viewing 8 posts - 1 through 8 (of 8 total)
  • Author
    Posts
  • #44249

    Split1294
    Member

    As a Black Belt I have been tasked with a variety of improvement projects.  However, one in particular is proving to be difficult to get my hands around. Customer complaints as a result of shortages, product damage, and ‘shipped too’ errors.  I am working with a team that has been selected by me, which is cross functional, Customer service reps, etc. I decided to use the FMEA tool, because we do not have a lot of historical data on errors. Moreover, I have been asked to work on preventing additional shipping errors from reaching some of our most important customers. The FMEA seems to be the logical tool.  I need advice, how do I maximize the effectiveness of the tool to get cooperation from; the IT department, Customer service, etc.?

    0
    #141488

    diboy
    Participant

    I work in a transactional service environment on our Six Sigma team. (MBB) Just my opinion but I would start by gathering internal and external VOC and follow up with hard data on the types of errors that are occurring. Then use some graphical tools to stratify your biggest pain areas such as a Pareto Chart. You may have to collect the data manually which means you will need to get the buy in from your different stakeholders to spend the time necessary to collect meaningful sample sizes. While you gather data take the time to process map the entire work flow so that as you see the defects, you can identify where they are happening.
    I have always used a FMEA at the end of the project when you have solutions to document what can go wrong and where it might happen after implentation.

    0
    #141503

    Clark
    Participant

    FMEA is an excellent tool in preventing errors. To make it work its best you need first to establish a cross functional team inclusive of IT, among others. Map your defined process then identify what can go wrong on each activity and challenge your team to identify on your end, e.g. what can go wrong on activity a based on the defined output as enabled by IT.
    You can also partner using Fault Tree Analysis to FMEA to dig deeper in your analysis

    0
    #141733

    Deanb
    Participant

    FMEA is the logical tool for 2 reasons: one because you do not have reams of historical data (but do probably have people knowledgeable of the situation and related processes), and second, you need various functions to learn from eachother so they can cooperate to prevent the failures. The FMEA may point you to where to start collecting data. You have a low occurrence but potentially high impact failure modes, which you need to identify, organize around, and mitigate. This is precisely what FMEA is best at.

    0
    #141734

    Split
    Member

    I could not agree more with the context of your message.  Justifying proactive measures which require time and money are hard to sell. Hopefully the FMEA project will validate the proactive measures our team will recommend.  Creating the sense of urgency is the key.
    Thanks for your input.

    0
    #141735

    Deanb
    Participant

    Split,In such cases I place a high premium on consensus and people knowledge over data knowledge. Granted, it helps to have some data upfront to engage the FMEA teams, however this initial data is rarely conclusive by itself. You need the involved parties to explain how this data is happening within their processes. I have found that getting this step zero right is the key to getting results down the line. At this staqe I would only propose investing in data gathering if it contributes to the FMEA facilitation or to consensus building.

    0
    #141741

    Stevo
    Member

    It’s not uncommon to have a lack of direct VOC, however with the right subject matter experts, you can get pretty close. I would start with a simple fishbone, see if there are any trends and take a deeper dive where needed. 
    I understand time and money are at a premium these days, use as many tools as needed (within timeframe and budget), but don’t be afraid of making decisions without 100% of the data.
    Stevo
     

    0
    #141758

    villageidiot
    Member

    You gotta do two things:

    Determine what defects are occuring, where, and their impact on the business (costs) and the customer (customer satisfaction=market share)
    Determine the cost of this poor quality (COPQ) as it pertains to these internal failures (defects before you ship) and external failures (defects after you ship)
    You need to identify the defect types, create defect  codes and way to quantify their rates of occurance.  Prior advice is fine:  p-map & FMEA combo works well, but a quick affinity diagrams & IRD might introduce a preliminary cause and effect relationship that will help determine where to begin your efforts….
    All work well, but I would want to validate the pareto using real data.  This shouldn’t be too tough…for external failures, look for reactive sources such as warranty claims, field visits, etc.  For internal failure rates, have the operators capture defects using check sheets anywhere value is being added in your process (process steps, work elements, etc).  Use DPU here, convert to RTY using inverse natural log of DPU, and then calculate your cost of nonconfornance, or CONC.  Use accounting and pareto out the dollar cost of poor quality.
    Once you have your defects quantified and their effects in both hard costs (internal failures) and customer satisfaction (external failures), refer to management as to their preference:  Do they value VOC over cost or vice versa?  Now you know where to start.
    Once you have a good idea of defect frequency, you can calculate out your RPN.  I would suggest defining your “severity” scale based on some form of customer value propositions (ie 10 in severity means it directly impacts a highly valued customer….don’t forget to define “highly valued”….eg high volume, high margin, highly stable shipment history…. etc).
    Good luck.

    0
Viewing 8 posts - 1 through 8 (of 8 total)

The forum ‘General’ is closed to new topics and replies.