A Clear View Is Best
A customer was sharing a challenge she faces each day at her company. She is one of three managers at a packaging manufacturer that plans and schedules production runs. Here is the problem as she sees it:
“My schedule is based off order data that comes from our accounting system. We transfer new orders to our planning software each day by manually keying in the data that helps us track the important information we need to schedule people, machines, and materials. As soon as I commit the schedule to the system it is incorrect. I know it’s incorrect and so does everyone else. If I could somehow get the schedule visualized in a way that could make sense to the rest of my team I think we could improve our process.”
Wrong Wrong Wrong
If you visualize incorrect data then really all you’ve done is just beautified the lie.
My question to her was why is the data lying to you? Here is what we discovered.
Treating everything as a task is the simplest way to explain the problem. Yes you have orders. Yes you have quantities. Yes you schedule people and machines and materials. But when everything is scheduled what really matters most and how do you track the changes that occur?
We experimented with our theory for one week. Simple tasks assigned to a specific person. Each person assigned to a machine. Tasks, person, and machine connected to an order. Order was connected to the delivery date.
Here is what we found to be true – (1.) The closer you can get to the point of production (meaning the individual person responsible for the task) the more accurate the data. (2.) The closest you can get to near real time feedback the more accurate the data.
So what happened during our experiment that impacted my friend?
3 tasks were assigned to one of her shop floor workers – Start order #320014, Make 150,000 things, Finish order #320014, and she assigned the worker to a specific machine #13.
The worker started the first task, then the second task. Two times during the day machine #13 broke down because of a jammed head. Once the worker was able to fix it in 20 minutes, but the other time maintenance needed to be called to fix the head. The second time the machine was down for 2 hours.
The minute the machine went down the worker changed the status of his task from active to stopped. Entered in a reason for the stoppage and an alert was sent to my friend letting her know that the schedule was creeping. 20 minutes the first time. 2 hours the second time.
What did that provide her?
She knew that there was no way that order #320014 was going to be ready the next day for shipping so she was able to call two other plants to see if they had any extra inventory to complete the order on time. One plant had enough to fill the balance of the order and put the product on a truck that was leaving at 4:00pm that day.
Before our experiment my friend wouldn’t have found out the order wasn’t going to be late until the next day’s production meeting. Simply put, progressing the tasks and communicating change to the schedule provided her the clear view of the truth she needed to make a smart business decision and make the delivery date promised to the customer.
If you want to learn how we can help you see the truth in your data clearly give us a call or send us a note at firstname.lastname@example.org