Managing production strategy - the way forward

Read so far

Managing production strategy - the way forward

May 12, 2014 - 19:59

Fig. 1 - Mill and paper machine selection during order entry

BRUSSELS, May 1, 2014 (PPI Magazine) - "Life is short when changing habits deep in culture, organizations and human behaviour".

Changes in the market situation have dramatically changed the way paper industry works today. Instead of solely trying to reach profitability by producing volume, today's survival strategy in production is: Always keep your promise to the customer, at the same time reduce unnecessary working capital and optimize production cost.

Production is expected to produce exactly as planned, both amount and time wise. However, as the nature of production does not allow exact production, as in the manufacturing industry, new thinking and tools are needed to manage any under/over production situation. This is needed to create a smooth flow of products and avoid unnecessary rework after production.

The right amount at the right time

Major organizations are moving towards cost-based planning, where production units inside one organization compete for customer orders. During order entry, orders are not directed toward a specific production unit but the order goes to the mill which is able to produce and deliver the goods in the requested time with the lowest cost.

This means that to receive an order in a modern supply chain, each production unit must keep production costs as low as possible and provide maximum flexibility. As individual mills cannot easily influence raw material or transportation costs, they need to concentrate on

  • Reducing working capital
  • Reduce any extra operation costs such re-work costs
  • Increase cost awareness by providing immediate feedback of production costs during manufacturing
  • Keep production plans as flexible as possible to allow late changes in scheduling for new urgent, highly profitable orders.

Luckily these objectives are not contradictory and can be achieved with business process changes in production planning and processes. The main target is to produce as late as possible and produce exactly what is needed.

Reasons for over/under production

It is very typical in the industry for the final production numbers to be different from the original order. This happens for several reasons:

  • The order tolerance is not totally clear for sales, planners and production.
  • Seasonal variations influence length and diameter calculations which are not considered in ERP systems during order entry.
  • Production does not always follow a planned diameter but still follows a planned number of sets. This is one typical reason for underproduction.
  • Winder operators only see the need for additional production too late, after the paper machine has already moved on to the next product. As a result, orders are under produced and the planner needs to take corrective action (rework).
  • The paper machine tends to produce too much to compensate for possible waste in later production phases. This is a the typical reason for overproduction.
  • Overproduction from a paper machine cannot be easily assigned to new orders, thus operators tend to do what humans always do: Follow the easiest path and simply overproduce already planned orders.

Typically, organizations react to over/under production by activating several programs which mainly try to get rid of already produced goods.

  • Continuous weekly follow-up of old stock situation in the supply chain
  • Use inventory teams to find target customers for free and old stock
  • Activate production planners to utilize old stock for new orders.

Still, organizations seldom really try to overcome the root cause, mainly because existing Mill Execution Systems (MES) lack the proper tools to guide work in the paper machine and winder where major production decisions are done.

Prevent over/under production

Starting with overproduction, the first goal in the battle against it is to make it visible as soon as possible. Most MES systems do have a function to identify overproduction and set it free before or during wrapping. This creates immediately important benefits:

  • Overproduction is immediately visible for production planners to be used for new orders
  • Warehouse users will not store overproduction in the same location with original customer order (business rule).
  • Overproduction is not accidentally sent to supply chain where relabeling and rewinder capacity is not easily available

Separate storing of overproduction will create extra work in warehouse, as operators need to find a new location for it or, alternatively, use clumsy mix locations. As a result we get immediate feedback; warehouse users will start complaining of overproduction to production operators, which clearly increases problem priority. A dedicated overproduction area in warehouse will also make problem easily visible to anybody walking around a mill warehouse.

The next step is to start influencing the root cause: why overproduction happens. As we have seen many times this is not always an easy task because of the nature of paper production. Machine chains can be long and produced waste is hard to predict. Finally, a long tradition exists in business to optimize pure paper machine production. The focus need to be set on the following improvements:

  • The paper machine will produce the correct size of jumbo reels taking into consideration the winder sets
  • Winder production target is to fulfil orders inside order tolerances.

Influencing paper machine production is technically an easier task. Many mills still have a habit of always producing to a paper machine's maximum diameter. This makes operators less aware of final target: customer orders which are produced on the winder. In an extreme case when the winder operator is asking for more paper, the machine produces one more jumbo reel, when a small adjustment would have been enough.

In our target situation, where the paper machine produces winder sets, produced jumbo reels do fit better to winder patterns and winder personnel have a common language when communicating with the paper machine. The MES needs to provide operators with an easy tool to see the optimal size of the next jumbo reel and a clear means as to how to run the machine to achieve optimal size.

The winder operator needs to cut production from the paper machine to actual customer orders. Often this is a simple task - simply follow the production plan -- but in under/over production situations, it becomes difficult. A key process here is the order selector algorithm, which is commonly used in a modern paper mill's MES. However, new tools and algorithms are clearly needed to better manage various production situations. The following major rules and tools are needed:

Yield calculation: The winder operator who is responsible for order fulfilment needs an early warning that an under/over production situation is expected. Completed production is compared with plans, future yield is evaluated and the operator is given a clear indication what the end-result will be if production continues as planned. This makes it possible for the winder operator to discuss with the machine operators early enough if more (or less) production is needed.

Planned versus ordered amounts: The winder operator has two different operating modes. For orders which are not completed in the current production run, the operator simply follows planned production amounts. For orders which will be completed, the operator's task is to make sure that order will be produced within customer given tolerances.

Use rewinder: In the case of a minor underproduction situation, one or two reels short, the winder operator can ask for support from the rewinder process. The rewinder operator has a mandate to search suitable, free reels from the warehouse and re-assign or cut to new orders. A good control of warehouse and warehouse reservations is needed to avoid situations where the rewinder operator steals reels from the production planner.

Find target orders for additional production: In case of over production, e.g. an extra set is produced, new tools are needed to support the winder operator. Many systems already support re-trimming but operators seldom use those tools as the process is considered too complicated. The main problem is not just how to use the tool but also how to find suitable orders for additional production. This is a problem especially outside office hours when production planner is not available for support.

A new approach to the problem has been proposed. In a case where new orders are needed, the system runs a cost-based order selection algorithm and proposes the best fit for extra production. When the operator accepts the result, the system makes an automatic re-trim. The algorithm favors orders from the existing production run as a first priority. It is important that production planners have the flexibility to assign extra orders into a run to be used for re-trim cases.

Using orders outside of existing production run is not always a good solution, as it means pre-production, which in turn consumes warehouse space and makes late order changes more difficult. Therefore, a predictive business process to ensure proper production to fulfil orders at winder is always the first priority.

What to do next: If over/under production is as subject of concern, how should the mill approach the problem?

The first thing is to measure the existing process by key performance indicators using clear, mill-independent calculation rules. This data can be used to compare mills and to start understanding to which extent a problem exists. KPI's are calculated for each PM separately.

The main KPI's to consider here are:

  • Amount of free stock in warehouse
  • Amount of stock which is older than a set number of days in the warehouse (tons)
  • Over and under produced tons (compared with tolerances) relative to planned production volume (tons and %)
  • How much overproduced volume has been used (reworked) for other orders after wrapping (tons and %)
  • How much over/under produced volume has been sent to a customer or to an external warehouses (tons and %)
  • Tons sent to customer a set number of days after or before latest mill ready date (tons and % of delivered amount)

Based on these figures, an idea if over/under production is a problem on some production lines should already exist. Sometimes high numbers can be easily explained with business reasons, like repeatable business. Sometimes actions are needed.

The first step after finding an issue is to check that production planners, machine operators and production supervisors work together and understand each other. One good approach is for a production planner to move his/her working place to the winder operator's room for some period, e.g. one day per week, each time on a different machine. People then learn each other's tasks and start understanding why people work as they do. This kind of process can be a very eye-opening experience for both parties.

As the process goes forward, the next step is to create a development program with a clear aim of reducing working capital and making production more agile by producing correct amounts as close to the planned delivery dates as possible.

This includes continuous and visible KPI follow-up, training sessions for key operators and often also changes to existing operating systems. New system support to calculate optimal PM jumbo reel size and early warnings of possible under/over production situations are essential requirements for a modern MES system.

Antti Blomqvist is Principal Business Consultant at Tieto with many years' experience with the forest products industry.

Pulp & Paper International is FREE to qualified subscribers.Click here to find out more.