With several thousand products and business expanding,Finnforest UK needs to produce highly accurate forecasts.SAP Advanced Planner and Optimizer (SAP APO) enables them to do just that.The key to the success of their SAP APO project so far has been the comprehensive planning stage before starting the actual implementation.

Forecasting buying and selling in any industry is a very complicated process. In the wood products industry, where Finnforest UK does business, forecasts are crucial because of the thousands of products the company sells, the multiple sectors it serves, and the relatively low margins in the industry. The Finnforest business in the United Kingdom (UK) is its largest outside Finland. It includes raw materials coming from Finnish forests and sawmills the company owns, with finished goods manufactured and distributed in the UK. The companys business model has become ever more complex and is growing in the four distinct areas it serves: do-it-yourself (DIY), merchant business, industrial business, and building and construction, with a focus on the UK and Ireland.

As Mark Tonge, director of logistics and information technology, recalls, they used to have a very primitive way of dealing with forecasting, using spreadsheets. The spreadsheets had evolved over many years, some were out-of-date, and they were very prone to error, because it was an entirely manual process. The spreadsheet figures were entered manually into Finnforests SAP R/3 4.0B system. Tonge emphasized that because of this they were spending an inordinate amount of time on input and little on analysis. And the all-important mean average percentage of error (MAPE) was just too high to be usable in the business. They realized that this solution would never be sustainable over the years, or robust. And not all areas of the business were even forecasting. What they needed was to standardize their forecasting processes across the business.

Why they chose SAP APO

So the company began looking at integrated forecasting systems in a joint effort between the business and IT sectors, mainly in the supply chain and logistics areas. As Finnforest UK recognizes, those areas are the most affected when forecasting does not work or is not accurate enough. After thoroughly analyzing what was available on the market, they chose the demand planning functions in SAP Advanced Planner and Optimizer (SAP APO), mainly because were very committed to SAP with our R/3 ERP systems, notes Tonge, and the fact that they are interested in implementing other parts of SAP APO later. Also, they felt that the SAP APO algorithmic detail and user interface were very good.

Their main goal in implementing SAP APO was to significantly reduce MAPE even down to individual product or stock keeping unit (SKU) levels. But Tonge also envisions expanding the use of forecasting throughout the business to include supporting the budget process on the commercial side, which they will probably do for 2004. They had been hoping to do this for budgeting for 2003, but the timeframes were not compatible. Another goal is to make reviewing these forecasts a part of each monthly board meeting.

A main driver of these combined forecasts will be Crystal Reports from Crystal Decisions, which they already use for many of their reports across the company. In the Crystal Reports they will be able to combine data from SAP APO and SAP R/3 together with Finnforest financial and logistics data. Currently, they are using the Business Explorer included in SAP APO and are looking at how best to use Crystal. Because they employ an exception-based analysis bias, their reporting will be customized to focus on exceptions.

Factors for success

Getting down to business, the implementation project was largely driven by Finnforest themselves. The first phase of the project was implementing demand planning in the DIY stream of the business. The second phase, which is currently being completed, has seen the rollout to the other three streams, so that SAP APO is used throughout Finnforest UK. With SAP APO, Finnforest has been able to improve forecasting accuracy dramatically for around 5,000 SKUs and 4,000 delivery points. According to Vinod J. Thayil, logistics business analyst, they should see a reduction of up to 20% in forecasting error. As Tonge says, Its a whole learning process for the entire organization.

From previous experience, they recognized that to make sure the project was a success they needed to own it. This had not happened in previous projects, which had hurt Finnforest in terms of flexibility and support. Working with consultants from SAP UK, Finnforest UK spent several months planning every detail of the project. Good planning will result in an effective and hassle-free implementation, as James Jarvis, IT business analyst, emphasizes. Preparation meant spending a lot of time planning the integration with their SAP R/3 system, assessing the key performance indicators (KPIs), and more. They did not try to rush to implement the SAP software; they wanted to put the proper foundation in place and go at an appropriate pace without a definite deadline. And transfer of knowledge was key. Thayil worked directly with the SAP consultants, whom he found to be very good.

Whole organization behind the project

Another important factor for success was that the organization as a whole was behind the project, because otherwise your risk of not making the project work is that much greater, Tonge notes. They spent a large amount of time making sure everyone was behind the system, meeting personally with all the employees who would be involved in using it. That has really paid off because everyone that we have demonstrated to can see the benefits of where we are going with this, Jarvis says.

A dedicated company forecasting resource makes sure that sales people update forecasts and acts as a focal point for doing the analysis. That is how they want to go forward with the SAP APO solution. They have even re-designed some business processes to make sure the forecasting and SAP APO work optimally. Currently, the system has 10 users, including the account managers. And user feedback has been very good both in terms of quality of the forecasts and ease of use of the system. User acceptance is crucial because, as Tonge puts it, forecasting has a touch of art as well as being a science. They are currently in the training process, adding users in finance, marketing, and even the directors of the company.

Excellent implementation model

Their model has been very successful, and Tonge goes so far as to say that this implementation has been our best experience with SAP and our best experience with anyone. They will definitely be keeping this implementation model for future projects. One current focus is on the material requirements planning (MRP) functions in SAP, which is going to be a bigger project. There are many implications in running an MRP system, Jarvis emphasizes. They have already set up a working group to look into how they can improve use of MRP functionality.

With SAP APO in place at Finnforest UK, the company is looking to roll out the system to subsidiaries in France and Germany, though not before 2004. They are currently looking at introducing collaborative forecasting and have already met with one of their customers to gauge interest. They would also like to share data back to Finland, increasing the transparency and visibility for determining raw materials requirements.

Collaborative forecasting would work through an extranet with secure logons. With foresight, they had already planned for this during their SAP APO implementation and added a field for customer forecasts. For Tonge, this is important, because the closer you collaborate and cooperate in our industry, the bigger the value and the gains for the parties involved. Collaborative forecasting can provide a comparison of numbers from the customer, the system, the sales manager, and the central planning resource to provide several perspectives and reference points prior to finalizing the forecast.

Other projects include an upgrade of their SAP R/3 4.0B system by the end of 2004, because of maintenance ending for that release; an electronic warehousing project, implemented in conjunction with an SAP partner; and an Internet sales project involving Finnforest UK and SAP Finland. All of these and future projects will focus on the forecasts the company generates. The forecast is a key driver in anything they do, whether looking at MRP or further supply chain solutions. And as Jarvis says: Its actually pretty exciting because we are just at the start.

FINNFOREST CORPORATION

Finnforest Corporation, owned by the Metsliitto Group, is Europe's biggest international mechanical wood group with revenues of 1.7 billion (U.S.$1.8 billion). Moelven Industrier ASA is Finnforest's majority-owned Norwegian-based subsidiary. Operations are grouped into two industrial divisions: Engineered Wood and Solid Wood. Finnforest's main geographical markets are Market West, Market East, Scandinavia, and Finland. Its customers include industrial end users, DIY stores, building suppliers and timber merchants, building contractors and developers. Finnforest Corporation operates in over 20 countries and employs over 7,500 people.

Customer Success Story appears in SAP INFO Issue 102

Please click here to read the success story in PDF format.