Real Time Value Stream Data Collection

 

Value streams require real time data collection in order to control them. This is often done automatically by the equipment itself but in some occasions it must be done manually. 

In spite of the high-tech of our shop floors you can still see pencil and paper data collection being used.

https://youtu.be/C7chwW_cXbc

Unfortunately, detecting aspect defects is often difficult to automate even with AOI (Automatic Optical Inspection).

When an operator is required to inspect we need a method to enter the results into the system.

Line incidents, change of critical process parameters, etc. also need to be reported the moment they take place.

Pencil & Paper Reporting

The method still widely used in this 21st century is:

1. The operator hand writes the defect data on a form
2. Forms are collected at the end of the shift
3. A clerk types the forms contents into the system at a later stage
4. The engineer analyses the results and takes corrective actions well after the data was entered.

This reporting process creates a number of problems:

• It is time consuming for the operator
• Transcribing the data is a non-value-add operation
• The operator’s handwriting is not necessarily good (this criteria is not used for operator selection)
• Reporting errors may be introduced both by the operator and the clerk doing the transcription
• Some of these errors may be impossible to recover when they are detected on transcription
• Real time process control is impossible: the engineer detects a problem at least 24h after it happened.
• Maybe they have been producing bad product all this time.

Decimal Point Responsible for Death

In a large hospital a doctor's prescription was misinterpreted by the hospital pharmacy due to the missing (or not clear) decimal point in the dose. 

As a result a child was administered a dose ten times the intended value by the nurse and the result was death. 

There were, possibly, several other mistakes by the pharmacy, nurse, etc. who did not detect the mistake. 

But this an example of a terrible consequence of misinterpreting hand written information. 

These sort of hand writing mistakes are very common in all sorts of value streams. 

On-line Reporting

The alternative to the pencil & paper data collection is, of course, directly reporting incidents in a terminal the moment they happen.

You can provide terminals in the inspection and test stations to do that.

Sometimes it is difficult to provide individual terminals to all the operators who need to report. 

Sharing terminals and having to walk to the nearest terminal is time consuming so it may not be acceptable.

The reporting operation should not be time consuming and error prone. 

If we ask operators to type 12 digit numbers or timestamps the chances of error are very high.

So we need a reporting system which is both simple and error free.

Bar or QR Coding

Large number typing can be replaced by a bar code generated with Office and read by an APP in a smartphone:  

Another alternative is a QR code which can be automatically generated with an APP. 

Data Entry Validation

We can use the data validation alternatives provided by Excel:

In this way the reporting errors may be detected the moment they are entered and corrected on the spot.

Pull Down Menu

Pull down menus may be used for each field to avoid typing and insure all use the same wording:

Time Stamp

The current time stamp can be automatically recorded with a formula:   =now()

Bank Account IBAN Check Digits Validation

Critical numbers such as a bank account IBAN number can be easily validated with a formula the moment it is entered so that it is immediately rejected if the control code is wrong.

Reporting on the Cloud

Having a terminal for each operator who needs reporting may not be possible, in which case reporting on the cloud with the operator's own smartphone may be an alternative.

 
The data base can be created in a spreadsheet in Google Drive, One Drive or company web :

Reporting may be done with any tablet or smart phone via WIFI

BYOD (Bring Your Own Device) Alternative

Operators may have difficulties with readable hand-writing or reporting in a terminal but they are normally experts in typing into their own smart phones. This existing skill can prove to be very useful for real-time on-line reporting.

 

When a problem is detected, it can, immediately, be reported with the phone from anywhere and the moment the problem is resolved this can also be reported on the spot.

The smart phones used may be the operator’s own device or one supplied by the company.

It, obviously, requires appropriate firewalls to avoid entering a virus.

Even a smart phone with no telephone line may be used to access via WIFI (an old out-of-use smart phone might do the job).

To avoid corrupting the database each workplace may have a dedicated input worksheet in the cloud and then data is automatically transferred to a central worksheet. 

Operators need write access to their input worksheet and only read access to the central worksheet and reports.

Voice Data Entry

Data (both text and numbers) can also be entered by voice with an ordinary smartphone (Android):

Real Time Operator Feedback: Weight Control

To control your weight you can use this Excel SPC in the cloud with your smartphone:

 

You can easily enter your weight and immediately know if there is a significant trend up or down. Apart from checking if your weight has gone beyond the upper or lower control limits other checks for trend symptoms are applied. In this case looking at the graph is not really necessary for your real-time control.

Real-time Feedback to the Workplace

Real-time data collection enables real-time feedback to the operators in order to take corrective actions while it is still possible.

This Andon display provides real-time feedback:

• PTH component insertion DPMO (red means above 1500 target)
• PTH component solder DPMO (green: below 190 target)
• Processed cards/hour.

SPC (Statistical Process Control) Charts 

SPC Charts are another alternative to provide feedback to the operators in real-time.

When this feedback is in real-time it is possible to relate changes in the output to what just happened in the line: change in process parameters, line incidents, etc.

This makes it possible to get to the root cause of problems and often solve them on the spot.

Project Management Data Entry

In complex projects where multiple people and organizations are involved it is essential to keep  up to date everyone involved in real time. 

Since jobs are interdependent any delay can affect several jobs downstream.

They all need to know when the preceding jobs are completed to plan their own. 

Being able to access the updated schedule anytime from anywhere is easy when the project schedule is in the cloud and accessible by all participants with their smartphones.

Conclusions

• Pencil and paper data collection makes it impossible to provide real-time feedback to the operators: all we have is after-the-fact compound data. 
• In this case it will be difficult to find the root cause of the problem and therefore the problem will often remain unresolved. 
• All we can do in this situation is guess what happened and try to find a justification for management.
• With pencil and paper data is often missing and there are errors which often can't be recovered.
• Data entry errors if not detected in the moment of entry may be impossible to recover.
• Data may be easily entered into a spreadsheet in the cloud, via WIFI, using an existing smartphone.
• Direct data entry into the system enables automatic data validation which will request the operator to enter again in case of a typing error.
• This data collection alternative may be used to build a Value Stream Map while visiting the line.
• It is essential for Real Time Statistical Process Control
• This remote access can be used  to update or view charts of a Virtual Obeya Room 
• Design of Experiments data collection with a smartphone via WIFI can be more reliable