Should you be Demanding SAP IBP?

Posted by Ian Brister on 05-Jul-2018 15:06:00

It has been a while since I first published this Blog. In the 2+ years that have gone by, a lot of functionality has been delivered with each upgraded version of SAP Integrated Business Planning (IBP), especially with the Demand application. 

In the world of ‘Internet of Things’ and Digital Transformation, where companies want to respond faster to customer needs, we have all read about SAP’s vision to enable businesses to transform from what were previously Supply Chains into Demand Networks.

But where does SAP Integrated Business Planning (IBP) for Demand fit in to this?  And as a Planner, should you be Demanding IBP?

In order to remove some of the uncertainty and puzzlement as to what exactly IBP for Demand can offer, in this post I intend to explain some key differences between SAP Advanced Planning & Optimisation (APO) and SAP IBP for Demand.

For the Demand Planning community, it is fair to say to that there have still been no major developments in SAP APO since SCM 5.0 was released. The statistical algorithms/functionality have remained largely the same (have you ever used Re-Initialization!), and the user interface experience in both planning and alert management has fallen behind the expected standards in the 21st Century.  Yet, Demand Planners are expected to forecast more Product/Location combinations with normally only a token understanding of what statistics is all about.

Released as part of IBP v5.0, the IBP for Demand application is SAP’s next generation forecasting application. Since SAP IBP was developed natively on the SAP HANA platform, its performance outpaces that of systems that have merely been modified to use in-memory technology.

ibp

The Fiori based web access to SAP IBP gives the Planner a seamless, responsive and cross device user experience (especially in Dashboarding & Analytics), whilst the MS Excel based planning interface puts users back in to their comfort zone of spreadsheets.

ibp2

But what is IBP for Demand?

A basic definition would be;

IBP for Demand = Traditional Planning + Demand Sensing

As SAP deliver more sophisticated algorithms over time, an enhanced definition would be;

IBP for Demand = Traditional Planning + Demand Sensing + Machine Learning

          ARIMA

Statistical Forecasting

Prior to the execution of the statistical forecast, SAP IBP for Demand performs pre-processing steps;

  • Substitution of missing values
  • Outlier Correction (with interquartile range and variance tests)

The missing values step is not part of SAP APO, but what SAP IBP allows is for a pre-run with just the outlier correction step. This means that you could have a different key figure to store the system proposed correction, and allow for Planners to check prior to generating the actual statistical forecast - or use alert functionality to only review those corrections outside a certain tolerance.

Since the release of 1802, SAP IBP is now head and shoulders above SAPO APO when it comes to the range of univariate statistical forecasting algorithms available used to derive the medium term forecast. SAP IBP 1802 delivered;

  • ARIMA (autoregressive integrated moving average)
  • SARIMA (seasonal version of ARIMA)
  • Brown’s linear exponential smoothing (lessens the trend effect similar to double exponential smoothing, but using a single coefficient to smooth both the level and the trend in the historical data)

and IBP 1805 improved the Croston method.

Machine Learning

"SAP offers comprehensive data science tools to build models, but it is also the biggest enterprise application company on the planet. This puts SAP in a unique position to create tools that allow business users with no data science knowledge to use data-scientist-created models in applications". Forrester

In the directional product overview below, on top of the Demand Sensing that is already available and in the next section, SAP should offer machine learning based algorithms to support areas such as;

  • Statistical Forecasting (with Gradient Boosting)
  • New Product Introduction - interesting to see a PoC with Customer Data
  • Segmentation analysis
  • Master Data
  • Exception Handling

ibp3

Demand Sensing

Obviously the major functional advancement compared to SAP APO is the inclusion of a Demand Sensing algorithm for the short-term.

ibp6

Within SAP IBP for Demand, Demand Sensing is available to run at Product/Location/Customer level only. A criticism of SAP would be that Demand Sensing still only supports a single downstream signal - something that is being improved on in 1811 to hopefully allow for other factors such as EPoS, weather etc. signals to be included.

Results from a Proof of Concept that compared the forecast accuracy using “Traditional DP” forecasting methods with the inclusion of short-term Demand Sensing within IBP for Demand, showed compelling improvements in forecast accuracy.

Traditional DP vs Demand Sensing

Planning with Microsoft Excel

The SAP Integrated Business Planning add-in for Microsoft Excel allows you to review and modify your planning data and run simulations.

Depending on the IBP applications that your company has licensed and configured, it also provides advanced planning functions, such as inventory optimisation, and statistical forecasting methods for creating accurate forecast data from historical sales figures.

In the view below, the power of SAP IBP and it's ability to allow users to Scenario plan shows a data set across both the Baseline and 'Blog Example' scenarios. Users can choose which User(s) or User Group(s) these scenarios are visible to - ideal where sensitive information may be at play.

ibp4

The Future?

Version 1808 should finally see the long awaited data realignment functionality being available within SAP IBP - probably the last major hurdle in convincing legacy SAP APO customers to make the switch to SAP IBP.

With cell 'fixing' now available within the MS Excel add-in, and when integrated notes functionality is added allowing comments to both be added and reported on, then the SAP IBP for Demand product is slick and very user friendly. Two main directional areas for SAP are;

Time Series AnalysisAutomatically analyse sales patterns in your data e.g. for trends and seasonality to streamline the forecasting process

Forecast AutomationFully automised forecasting process including forecast level optimisation (something that was available as an add-on in SAP APO (Note: 1366618))

which, referring back to the Forrester quote, this will "allow business users with no data science knowledge to use data-scientist-created models in applications".

Collaboration - Due to licensing restrictions, it was forbidden for external partners to login to your SAP IBP system. So where a customer could be providing a forecast, or making changes on the fly during account meetings, unless the data was provided via interface the only way to get the customer data would be for your own staff to re-key. Often duplicated and wasted effort. This was a missed opportunity for SAP IBP to get real traction in a digital and collaborative world. But that is changing!

ibp5

Although I have not seen any licensing information, the directional product overview of WebUI access for partners will also help customers release business value from their investment in SAP IBP quicker.

Should you be Demanding IBP?

So what has changed over the two plus years? Nothing. The two main questions to be answered are, “Why would I implement IBP for Demand?”, and, “Does it replace APO”?  Although APO continues to remain an important part of SAP’s SCM portfolio, and will be in mainstream maintenance until at least 2025, IBP for Demand is here, mature, and the SAP Demand Planning application of choice.

IBP for Demand is not simply a copy/paste of SAP APO on to a new technology platform.  SAP have taken learnings from the APO past, and in the drive towards real-time information, better analytics and simplification, not all of the previous APO attributes have been, or more importantly, will be duplicated.

For those customers currently thinking of upgrading/merging existing APO instances, then the advice here would definitely be to pause and to look at the benefits going straight to IBP for Demand would give. To give SAP credit for IBP for Demand and HANA technology, the technology does not feel like the early bleeding edge days of APO 3.0/3.1, with thousands of customers having already implemented SAP HANA solutions.

For those customers already on the latest version of APO, SAP IBP for Demand should still be on your Planning Strategy Roadmap in assessing the value add that Demand Sensing functionality within IBP can bring.  An initial stride in to IBP could be to leverage existing APO systems and processes to provide a feed in to IBP for Demand, whose output would then provide SNP (or other 3rd Party systems) a Demand Sensed forecast.  The ultimate goal would be to then fully migrate the APO DP solution on to IBP for Demand.

Migration of APO DP & SAP IBP demand

Developing a better understanding

Whether it be assisting with the creation of your IBP Roadmap, giving IBP awareness training to your key business users, or hosting a Proof of Concept on our internal IBP system, our team at Olivehorse can assist you on every step of your IBP journey, and I hope this has been a useful read.

Click here for a full list of Olivehorse IBP blogs.

Ian Brister

Demand Practice Lead - Olivehorse Consulting

Ask us about any SAP IBP queries you have or about our range of IBP services

Read more on: IBP, IBP for demand, Demand