I answer questions about the economic effect of machine planning in the Ural workshops.

I answer questions about the economic effect of machine planning in the Ural workshops.

I recently wrote about the calendar planning system for the NLMK plant in Kaluga. There were many questions in the comments about the economic effects of this decision and who ended up doing a better job of planning: the system or perhaps a worker with 20 years of experience. This post is sort of an extended response to questions and comments.

For those who are interested in finding out what kind of savings are provided by “machine” planning using the example of the NLMK-Ural rolling shop, I ask under cat.

This is rolling shop #4

The implementation of Calendar planning and scheduling – KPiG or “calendar” for short – at the plant in Kaluga turned out to be successful for us (KPiG was also done at the Lipetsk site, but technologically it is completely different there), and we replicated it at the Ural workshops of NLMK: electric steel smelting and two rental There, due to the complexity of the supply chain, the economic effect was more pronounced: the Ural workshops are located in three cities of the Sverdlovsk region at a distance of more than 70 km from each other and produce a huge range of products, plus production is limited by many factors.

As planned “on paper”

First there was Excel. And the production planners had Excel. At first, the whole month was planned in a consolidated manner. The planning specialist decided how many rental campaigns there would be, i.e. of changes in the diameter of the reinforcement and wire rod (reconfiguration of the equipment for this takes a lot of time) and what assortment (by length, strength class, etc.) will be made within this diameter. They also set aside time for planned preventive repairs. The number of working hours per day was multiplied by the productivity of a certain type of product and the number of tons per day was obtained. For example: (24 hours minus 1 hour of prevention) * 158 tons/hour = 3634 tons

And the following schedule was published for the month:

Next, another planner divided the daily volume (our 3,634 tons) into the lines of the variable-daily task. Here it is necessary to assess which workpiece is in stock, which orders should be made earlier and which later. Also a critical factor is the availability of cars or wagons for shipment – they are not always available. Inventory balances and how quickly the warehouse needs to be unloaded are also affected. It is difficult for a person to make such a plan on paper about 3-5 days in advance. Here’s how the daily shift tasks, made by hand, looked like:

The main disadvantages of manual planning:

  • inefficient loading of production, increase of scheduled and unscheduled downtime of equipment;

  • increase in idleness of moving vehicles while waiting for shipment;

  • increasing the yield of inappropriate and long-standing products, due to the detection of errors during planning;

  • a long process of issuing a task to the workshop, as a result of the appearance of non-ordered products in the warehouse of finished products;

  • lack of understanding of the terms of shipment of orders from customers or even disruption of terms;

  • stocking of the warehouse, both in semi-finished products and in finished products.

What happened to KPiG

Now a complete detailed plan of production and shipment is built immediately for the whole month. If it is a preliminary plan, the preliminary demand formed on the statistics of the past 6 months is taken, if it is specified, then real orders are taken. With each planning run (when the user simply clicks on the “calculate plan” button), the person immediately receives a consolidated report of the planned production companies with a set of KPIs for its evaluation, as well as a daily shift task (SSZ) for transferring it to the shop.

This is a consolidated monthly plan, a replacement for Excel

This is a variable daily task that can be generated for any date range

Moreover, when creating a version of the plan-graph from the MES system, the most current data on the balance and shipment of manufactured products are immediately pulled up.

Therefore, the first plus, in my opinion, is an increase in the transparency of planning.

The second plus is an increase in the efficiency of loading production units and logistics;

The third plus is prompt and more accurate collection of data on balances, shipments, finished production;

The fourth plus is the reduction of warehouse stocks of finished products and semi-finished products.

The fifth plus is an increase in the accuracy of planning and its horizon from 1 week to 1 month.

Previously, specialists received input data for calculating the plan by e-mail or by phone – the percentage of errors was quite high. The risk of error is now significantly reduced.

The sixth plus – you can track the stage at which each order is. Planning is non-verbal, and therefore, in a daily shift task, it is immediately visible for which consumer the products are being made. Knowing the specific date of order production made it easy to plan transportation.

Seventh plus – OTIF is now more than 95%. That is, more than 95% of orders go to consumers on time. And the total economic effect is approximately 100 million rubles per year. For comparison, in the Metiz workshop, where the KPiG system has not yet been implemented, the OTIF rate is about 75%-80%.

Here, production redistributions that are already covered by the Calendar Planning and Scheduling (CPS) system are shown in blue, and not yet covered in white.

What is OTIF?

OTIF (https://en.wikipedia.org/wiki/DIFOT) is a key indicator of supplier performance.

It is calculated as: shipped orders / orders that should have been shipped according to the plan * 100%.

I am also pleased with how the project is evaluated by “business” – head of the production and shipping department Andrii Bilonogov:

“It is very important that our joint development has high variability, it literally adapts to any production situations and external factors. For example, unscheduled repairs on the line or newly received orders from key customers. We note this in the program and immediately receive solutions that are beneficial to us, which will allow us to switch to the production of blanks of other grades of steel without losses and save all delivery dates and key orders. Accurate calculations help to plan the release of the required amount of products and to ship them on time and in the required quantities for the order.”

Conclusion

In the first article, I wrote about an optimized approach to rearranging production intervals. And it looked like a comparison of who is better: man vs machine. But in reality it is not possible to compare “head on” because the KPiG system profoundly changed the very approach to the planning methodology.

* For the slides and diagrams, thanks to Maria Skiba, business analyst of the production and shipping department of NLMK-Ural JSC

Related posts