Big Data + Cognitive Technologies = Supply Chain Excellence

When it comes to getting – and staying ahead in the digitalizing economy, leading companies know they need to think outside the box. And they are embracing new tools to unleash the power of Big Data and power ever greater and sophisticated capabilities across the organization that can take them to new heights.

The amount of data flowing through the average company is staggering. This includes everything from structured numerical data to unstructured text and video and even social postings.

To be of value, this data must be collected and stored in a way that it can be accessed and analysed. And this has led to the adoption of sophisticated data management systems and analysis tools. In the past data was created, stored, and managed in private on-premise systems that were frequently in silos of business functions. And as a result, data, and the potential insights they could unlock with collaboration, were constrained by physical and organisational walls. But things are changing. Many companies have embraced enterprise-wide solutions and are now moving into shared cloud based databases that enable them to share information both internally and externally and essentially get the right data to the right people in the right places at the right times to drive optimal business decisions and outcomes. Modern solutions such as SAP S/4 HANA, an in-memory, column-oriented, relational database management system, for instance, can store and retrieve data at lightening speeds and support the Big Data revolution.

This speed is crucial to the utility of data in informing real-time business decisions – the kind that enterprises looking to gain and maintain competitive advantage must make. Broadly the value of data can be defined along 4 Vs:

·      Volume: the more data we have, the more certain we can be about patterns we see

·      Velocity: the sooner and more frequently we get data, the more relevant it is

·      Variety: patterns in data from multiple types and sources produce a complete picture

·      Veracity: conclusions are only as good as the information they are based on

To create value from this data, insight must be extracted using advanced analytics, an exploding field including business intelligence, general analytics, machine learning, and artificial intelligence. The potential value of advanced analytics on procurement and spend analysis alone is projected to be $100-$200 Billion. This immense value is derived from accessing the insights and patterns hidden deep within Big Data that can provide companies with clarity into what is happening right now, predictions of future events, and recommendations for actions to stay ahead of risks. In short, it automates the laborious task of sifting through noise, to give professionals the type of information and options they truly seek: the useful type.

To get the best value out of Big Data, Business Analytics is broken down to three distinct steps that build on each other and generate value for specific tasks, as well as for subordinated goals in cognitive computing, machine learning and artificial intelligence.

·      Descriptive Analytics: The first function analyses data of past events to answer the question of ‘What has happened’. Big Data can now be turned into meaningful insights, such as relationships between individual events. It reveals the impact of past actions, like the effect of price changes on demand. However, descriptive analytics requires deep subject knowledge for the right interpretation and usage.

·      Predictive Analytics: Predictive analytics goes a step further, by using historical data to predict future events and case outcomes. Decision-making is simplified, by providing the likelihood of events. Employees see potential impacts of planned actions or predicted risks and incorporate this insight to proactively conduct business. For example, supplier selection can be enriched by the potential fulfilment rate, as well as external factors including likelihood of natural disasters and future market downturns. Predictive analytics is based on data mining and machine learning algorithms. These require a significant volume high quality data to create robust insights, making it the ideal application for Big Data.

·      Prescriptive Analytics: The ultimate last step of Business Analytics enables the automated identification of the best possible course of action, regarding a specific goal. Employees are supported through distinct options that speed up decision-making. For instance, contract management is supported by providing the optimal contract conditions for a specific business case. To use this potential correctly, an oversight of employees is required to verify the provided options. 

And when Business Analytics is combined with cognitive computing and artificial intelligence, business gets smarter.

Take the procurement function. Supplier risk management refers to the identification, evaluation and resolution of potential risk factors. These risk factors include threats for project fulfilment, like product quality issues, financial affairs of partners, or external factors, like weather or traffic conditions. The diversity of potential risk situations need to be monitored, analysed, and handled. Therefore, internal and external data is collected, pre-processed and analysed.

Starting with descriptive analytics, information of previous events is examined to understand the impact of individual events. In supplier risk management the performance of specific suppliers is analysed. Delays, product quality and the fulfilment of contracts are evaluated through scorecards. Predictive analytics uses these findings for calculating future supplier performances. External defects can have an enormous impact on the business. For instance, natural disasters, like tsunamis or earth quakes in the production area of your supplier could lead to a total failure in the supply chain. Through predictive analytics, employees get deeper information outside of scorecards to make justified and informed decisions. Prescriptive analytics carries these possibilities forward by generating automated suggestions for detected risks. Alternative suppliers with the required capacity and performance are shown to handle the emerging risk and ensure the security of supply and buyers can quickly and effectively diversify and resolve supplier risk. Cognitive computing completes the supplier risk management process by fully automated risk handling. The suggested alternative of prescriptive analytics is now automatically rescheduled with closely no human action. 

With each step in the process the management of insight becomes faster and more efficient. Instead of employees sifting through large reports to try and make sense of current business situations, they are provided with clear insight and can spend time formulating the best path forward. Technology will never replace common sense and the insight of seasoned employees. But in leveraging big data and technologies like AI, machine learning and cognitive capabilities to make sense of it, procurement professional can work smarter and faster and deliver their organizations to new worlds of excellence.

Marcell Vollmer is chief digital officer of SAP Ariba and the former chief procurement officer of SAP.

SAP is the market leader in business applications; and SAP Ariba is the world’s largest business network, linking together buyers and suppliers from more than 3 million companies in 190 countries.

Tags

#DigitalTransformation #future #technology #ArtificialIntelligence #AI #MachineLearning #ML #SupplyChain #BigData #CognitiveTechnology