Confirmation of load by "piling" and whether it is a feasible schedule by "piling
In MRP's load calculation function, the standard load (cycle time) by item is set for each line, and the number of minutes of load to be applied to the line is calculated according to the order quantity, and then the load is "piled up" on the day when the lead time (days) is shifted, and then it is compared with the line capacity per day to check the daily win/loss.
In order to find out if we can meet our delivery date if we bring forward (allocate earlier) the overflow of line capacity that was found as a result of the daily pileup, we break down the pileup, and the production scheduler automatically does this for us.
Collapsing the mountain" means that we put the tasks in order so as not to violate time constraints and create a schedule that is theoretically feasible and is a method that emphasizes the calendar (horizontal axis) of work schedule planning.
However, in order to create an optimized schedule, it is impossible for the scheduler to set the constraints of the production site without any omissions. If you can confirm that there are no delays, that's enough. The rest of the system is a practical way to operate, where you can set up a safety stock of several days' worth of inventory just in case.
Under the condition that there will be no delays in delivery, it is more realistic to make a list of work lots within the daily capacity and issue instructions that say, "Please digest this amount of work lots on each line today," and this is a method that emphasizes the vertical axis of resource capacity.
Overlap Method, Safety Stock and Lot Size Relationship
As a result of assigning SE on the left and SSEE on the right
The safety stock mentioned above is taken into account during replenishment order generation during order deployment, but the timing of replenishment order generation depends on the method of overlap between processes.
In order expansion, the automatic replenishment function generates the manufacturing orders for the shortage of work input orders as replenishment orders (children) and the manufacturing orders for the shortage of work input orders of the replenishment orders (children) as replenishment orders (grandchildren) while referring to the manufacturing BOM to fill the order list.
When the overlap between processes is ES (End-Start), the process in its own process can start only after the work in the previous process is completed, so the replenishment order is generated at the earliest possible time at the start of its own process, and if a shipment or input occurs during the work period of the previous process, there is a possibility that the safety stock may be cut. There is.
On the other hand, if the overlap method is SSEE (Start-Start End-End), the previous process overlaps with its own process under the condition that the previous process does not end before the time process, so replenishment orders are generated at any time during the previous process, which minimizes the possibility of running out of safety stock.
However, SSEE is based on the concept that the manufacturing lot size is one piece and goods flow in units of one piece, and the actual unit of goods flow at the factory is the manufacturing lot unit or pallet unit, so the overlap between processes called SSEE may not always be possible.
What you can do with the "Collapse" function
As mentioned above, it is impossible to reflect 100% of the constraints in the scheduler, so the accuracy of the schedule generated is not 100%.
However, by collapsing the mountain, it is possible to create a schedule that is theoretically possible to achieve with no delivery date delays, which cannot be handled by the production management system's MRP "pile-up" function, and by using SSEE as the method for overlapping processes, it is possible to generate multiple manufacturing orders with lot sizes in units of one piece, just as in single piece production. Even if you don't, theoretically it is possible to create a schedule that adheres to the safety stock as much as possible.