SCHEDULING | INTERNET OF THINGS

Top

To perform any target to actual comparison and output Key-Performance-Indicators (KPI) such as Overall Equipment Effectivity (OEE) it was necessary to integrate scheduling data with the solution.

Goal:

• Automatically import scheduling data from various sources into into a single data repository

On first sight this seemed to be a simple task as all information needed was present in the existing ERP solution that was used to print out the work order travelers. This included part-number, work-order, operation, scheduled start and end date and planned time for execution and setup. However, it quickly turned out that it was not enough to only work with data available in ERP as it simply was too inexact for doing comparisons for short periods down to single shifts or even hours. To be able to do that information from the department responsible for detailed planning also had to be included. In discrete manufacturing this typically is a more or less manual process where one, after assuring that everything needed to execute a work order at the machine is available, schedules the job for execution within a short period of time (shifts, days or weeks). This includes material, NC-program, tools and all paperwork (e.g. traveler, drawings) that accompanies the work piece at the shop floor.

Fortunately the customer already had chosen Oracle’s Manufacturing Operations Center (MOC) which is an ultra-flexible tool when it comes to bringing together data from different sources into a single data repository. In MOC scheduling data can be imported and updated by various techniques dependent on the source. On the one hand predefined adapters can be utilized to automatically pull in data from a broad range of existing ERP solutions or databases. On the other hand external programs can directly push data into MOC’s underlying database. If none of these methods can be used it is also possible to store text files (.csv) which contain the scheduling data in a predefined folder. MOC automatically reads the files in adjustable intervals and updates scheduling of the manufacturing resources.

To accomplish this task at the customer a combination of several methods was used. More static data like material, work-order or operation numbers was directly imported from ERP. For data from fine planning that may change from day to day initially the file based method was used. The responsible employees simply maintained an EXCEL spreadsheet that contained scheduling data for the most recent work orders. In case of changes the only requirement was to change and save the spreadsheet. To further reduce the effort the customer meanwhile uses a self-developed program that provides a dialog for this task. Instead of outputting a file the program pushes the data directly into MOC’s database.

Functions:

• Acquisition and transmission of scheduling data from various sources into a single data repository

Benefits:

• Enabled automatic collection of short-term (fine planning) and long-term (ERP) scheduling information
• Minimized implementation costs by using easily adaptable interfaces to existing solutions