Why are there Customer Complaints when the process is in control?

So.. here is what my question is describing.

There is a bank and a manager is now using a control chart for customer service time. The manager has set Upper Control Limit of his control chart as their service time limit. The manager finds that the process is well within the control limits of the control chart, but there are many customers complaining about service times which don't meet UCL.

I have to explain why it happens... and have totally no idea. What is your suggestion?

Re: Why are there customers' complaints when the process is in control?

It sounds to me that the Control Limit on the time was arbitrarily set and not a result of customer investigation or study. You can set a limit anywhere you want but if there is not data to back up the validity of this limit, it is pretty much worthless.

Re: Why are there customers' complaints when the process is in control?

The study was conducted according to the measurements taken every hour for 8 hours for three days. So the Upper Control Limit was set by a certain study... =(

Re: Why are there customers' complaints when the process is in control?

jaywaee said:

So.. here is what my question is describing.

There is a bank and a manager is now using a control chart for customer service time. The manager has set Upper Control Limit of his control chart as their service time limit. The manager finds that the process is well within the control limits of the control chart, but there are many customers complaining about service times which don't meet UCL.

I have to explain why it happens... and have totally no idea. What is your suggestion?

Click to expand...

When you set your control limit with your process capability, it will always look nice and greener for you as you can meet it. Does it meet the general banking norms and time expectations ? Are you at par with other banks performance ?
If the customers are also customers to other banks and they find it quicker there to transact, every customer with you will have a complaint.

Re: Why are there customers' complaints when the process is in control?

silentrunning said:

It sounds to me that the Control Limit on the time was arbitrarily set and not a result of customer investigation or study. You can set a limit anywhere you want but if there is not data to back up the validity of this limit, it is pretty much worthless.

Click to expand...

Looks like they need a specification limit - not a control limit - which only looks for special causes, not meeting requirements.

Re: Why are there customers' complaints when the process is in control?

jaywaee said:

The study was conducted according to the measurements taken every hour for 8 hours for three days. So the Upper Control Limit was set by a certain study... =(

Click to expand...

Still a sampling - and subject to sampling error. At any point between the sample the lines can get bogged down. How many complaints are there? Do they fit the premise of sampling error? If not, it goes back to specification versus control limit.

Re: Why are there customers' complaints when the process is in control?

jaywaee said:

So.. here is what my question is describing.

There is a bank and a manager is now using a control chart for customer service time. The manager has set Upper Control Limit of his control chart as their service time limit. The manager finds that the process is well within the control limits of the control chart, but there are many customers complaining about service times which don't meet UCL.

I have to explain why it happens... and have totally no idea. What is your suggestion?

Click to expand...

First thing to suggest is to go back through your notes! As part of the course your tutor (or study curriculum if you are doing self study) should have explianed the difference between a specification and process capability. The two are not linked but the relationship between the two will tell you how well the process is able to staisfy requirements.

I think of this as a 3 stage process:

1. Establish the requirements. These may be specified outside by a part designer or by law for example and outlines acceptable tolerance. For your example this is often decided by someone in marketing based on customer surveys - 'How long are you prepared to wait in a queue for services at your bank?'
2. Establish process capability. This is a sampling exercise to establish what the 'front of house' capability of the bank is: 'How long does it take people in the queue to get served. For this to be effective it needs to consider busy times of day as well as slack times, staffing rotas and all days of the week.
3. Compare the two. Based on your process study you will have identified a mean waiting time and a measure for variation. You can use these to compare the (wide) range of waiting time with what is acceptable to your customers.

My guess from the way you have phrased the question is that the examiner is looking to you to describe more about steps 1 and 3 in your answer.

Hope this helps.

Re: Why are there customers' complaints when the process is in control?

It is important to note the processes may indeed be stable and predictable, but performing miserably. Dr. Deming's Red Bead Experiment is just such an illustration. The process ALWAYS leads to the company being "fired" and closed down. But the data are "in control".

As suggested by previous responders, please be sure that the UCL has been set mathematically from the process results.

Given that the customers are complaining, and assuming the process is "in control" (stable and predictable), then if you want to reduce complaints and have (happier) customers, then you must fundamentally change the process.

Re: Why are there customers' complaints when the process is in control?

Paul Simpson said:

First thing to suggest is to go back through your notes! As part of the course your tutor (or study curriculum if you are doing self study) should have explained the difference between a specification and process capability. The two are not linked but the relationship between the two will tell you how well the process is able to satisfy requirements.

Click to expand...

Actually, process capability is related to specification - process control is not (at least in the processes where Shewhart charts apply. In other dependent variable cases, it can relate to specifications - but the curriculum will not likely recognize that.)

Re: Why are there customers' complaints when the process is in control?

Steve Prevette said:

It is important to note the processes may indeed be stable and predictable, but performing miserably. Dr. Deming's Red Bead Experiment is just such an illustration. The process ALWAYS leads to the company being "fired" and closed down. But the data are "in control".

As suggested by previous responders, please be sure that the UCL has been set mathematically from the process results.

Given that the customers are complaining, and assuming the process is "in control" (stable and predictable), then if you want to reduce complaints and have (happier) customers, then you must fundamentally change the process.

Click to expand...

Being in control, but not capable, is a bad thing. Bad specifications are a very bad thing (but not unheard of.)