iSixSigma

Performance measurement analysis

Six Sigma – iSixSigma Forums Old Forums General Performance measurement analysis

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • #49096

    Vallejo
    Participant

    I want to measure performance on a claim processing department.
    There is 40 employees divided in sub-departments of 5, composed by a team leader and 4 employees.
    Those are entry level jobs and the skills of all the employees are very similar.
    The tasks that the different employees perform are
    different , even into the same department , also some employees perform a variety of tasks.The way I am considering in approaching the matter is to time every task for every employee and figure out and time average per task.I will consider each minute a unit of time , in that way I can compare the performance of all the employees and know who are the ones that under perform and the ones that over perform.
    I am planning doing so using a weighted average for each activity.Employee from group 1 weighted value table Task Average time Weighted value
    Task 1 5 minutes 5 units
    Task 2 3 minutes 3 units
    Task 3 2 minutes 2 unitsEmployee from group 1 Points/Hour table Task Times task is performed Value Points
    Task 1 15 5 units 75
    Task 2 8 3 units 24
    Task 3 20 2 units 40
    Total Points 139The total points can be used as a performance measure , it also can be used as a comparison measure between employees who perform different tasks.Times Task Points
    17 Task 1 (7 units) 119
    21 Task 2 (2 units) 42
    Total Points 161Even though employee 1 and employee 2 perform totally different tasks , in can be said that employee 2 outperformed employee 1 by 15%.
    I designed this myself , and I don’t know where to find info on how to design a system like this , can you please give me any suggestions , is this the best approach? ThanksCarlos

    0
    #167187

    Deanb
    Participant

    Carlos,Lets say your study finds that some workers process claims faster than others. Does that tell you anything useful? What if slower is better for results? What are the business results that matter? Satisfied customers? The study should focus on learning about what matters most to the business. In a claims project I worked on we determined the most important output was 1) positive resolution from client’s perspective 2) information so deeper preventive efforts could be considered.

    0
    #167188

    Brandon
    Participant

    Carlos, Dean has given good advice again.
    As a general approach, unless you are absolutely certain, don’t pick a metric to start with; as you appear to be doing here. Identify your CTQ’s for the process. What are we trying to accomplish with this process and how is the customer impacted by the various activities?
    From that investigation you should be able to identify the Critical X’s – they may not be what you think they are at the front end. Once those are identified you can apply the DMAIC approach. Begin with the Problem Statement…that clearly defines the What, Where, How Much, What Impact, How Long.
    You may well not have to apply full DMAIC to solve the problem however I suggest you begin the front end of a project with the fundamentals EACH time. Assure yourself you are applying your efforts to the correct metric.

    0
    #167189

    Vallejo
    Participant

    DeanThanks for your info , yes customer service is important in some of the duties , (like calling).
    I worked on that department myself and I was one of the top performers not doing much , I used to spend a lot of time researching ways to systematize the forms used by the departments and still managed to be one of the best. (that is what my supervisor said , there is no way for her to measure it)so I know that there is a lot of slack and that is what I want to minimize.

    0
    #167191

    Deanb
    Participant

    Unless the ranking system you are planning has a direct end in itself-such as paying bonuses to the better producers, then your effort is still a means to improving performance in aggregate. After all, if you merely rank and fire/replace folks and the aggregate has not improved, your project’s value could be questioned. You need to improve processes in aggregate, not just improve isolated tasks or shuffle people.You seem to have deep knowledge of the processes under study. There is probably existing data on #claims processed by worker over a time period. I would consider starting with these to set baselines, then look for the easiest-best way to find the opportunities for improvement. The tasks you mentioned may not be the main X’s.Time studies are usually not the easiest-best way IMHO to start with. Just talking with people is much easier and might reveal some interesting things you have not considered, such as workers near the vents are freezing and waste time putting on sweaters, while those near the windows are roasting and wasting time shedding clothes and guzzling water. Or, perhaps some phones are defective, or a bully is disrupting others, or too many non-business calls are taken or made, etc. Point is, none of these are time killers your planned time study will measure. I like to start with the big picture and work backwards, not the other way around.What I like about the interpersonal approach is it also gives you get a chance to establish rapport with those you ultimately will be coaching on process improvements. Then after coaching, you can compare current performance with the baseline performance to validate improvement. Just an idea. Good luck.

    0
    #167203

    Deanb
    Participant

    For what it is worth, just talking with people in a friendly and non-threatening way about their work processes can be magical by itself. I have seen significant boosts in productivity just from talking to them, before any process initiative is undertaken!Let us know how your project works out.Dean

    0
Viewing 6 posts - 1 through 6 (of 6 total)

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