Digitalisation is the optimisation of processes and methods by the extensive use of data, made possible by computer-assisted use interfaces and high-performance software.
In order to make the best possible use of digitalisation, it is necessary to regularly evaluate the maturity of individual technologies, and to prioritise certain use cases.
Blockchain, robotics, prescriptive analytics, deep learning, artificial intelligence...
It has long been impossible to avoid these buzzwords in any quality discussion of digitalisation. Companies pursuing the vague aim of 'going digital' had long been faced with issues of use cases and implementation. However, after years of discussion, research and practical application, some technologies are now already so mature and established that they, step-by-step, are nearing the energy and commodities industry 4.0.
In order to remain profitable in a very competitive and rapidly changing market, companies can no longer avoid increased digitalisation of their business model. But what does digitalisation actually mean? We'll take the commonly used definition: Digitalisation is the optimisation of processes and methods by the extensive use of data, made possible by computer-assisted use interfaces and high-performance software.
In the following we will examine what this could mean for the energy and commodities market, using three specific sample applications.
Due to the multitude of systems and interfaces involved, back-office processes in energy and commodities companies offer considerable potential for optimisation as they currently require a large level of manual effort. In this regard the application of RPA, in the form of bots, i.e. programs that control other programs – or a new generation of 'virtual colleagues' – can provide an efficient solution. These can run in the background on a 24/7 basis and when predefined events occur, such as the receipt of an email, they can read automated data from the trading system, process data using Excel, write emails and much more. This creates means that all repetitive process steps that can be presented based on rules have the potential to be automated.
Popular candidates for the application of RPA are, for example, deal validation (involving the comparison of a transaction between trading systems and deal confirmation), payment receipt validation and specific regulatory reporting processes. Other use cases are comparisons between portfolios and ledgers, the triggering of payments with the occurrence of certain events or the automated recording of invoices following comparison with market data. All of these processes need to be carried out with care, and often also in real time. However, they are time consuming and add only little value and are therefore ideal for RPA, because the employees will be free to focus on value-adding activities once automation is complete. In addition, the set-up of RPA is normally a 'quick win' as it can be implemented quickly and its integration in the IT landscape does not need to be invasive.
Even though blockchain technology is on the way out of hype cycles, nobody would argue that it cannot be implemented extensively in the energy and commodities market in the distant future – in particular in relation to trade between decentralised prosumers. In this respect the first blockchain transaction between the two energy groups made possible by the Enerchain project at the end of 2017.
Advantages include direct trade without centralised intermediaries and the automation of necessary processes, in order to reflect the often very small amounts that are handled peer-to-peer in a trade-compliant way,
Blockchain could act as a single source of truth for the trade cycle, in which the transaction entry occurs via smart contracts, followed by direct deal validation, mark to market calculation and finally the settlement. It therefore offers the potential to automate the many manual processes in trading and to streamline the IT system landscape, which is currently often too varied.
It is difficult to say whether or not it will come to this, due to the many issues concerning performance, security and scalability that remain unsettled – but it is a possibility.
A challenge of liquidity planning in the energy and commodities market is the high number of potential factors that can influence future liquidity (interest rates, payment history, revenue from direct marketing, energy and commodities prices, uncleared margining items, weather data, etc.) Here the use of predictive analytics methods for the optimisation and automation of liquidity planning can help to create remedies.
The conditions required for implementation are the availability of detailed historical values (e.g. SAP liquidity analyser) and having bid data architecture in place (e.g. SAP predictive analytics for HANA). Once set up, the possible applications are very diverse. It is not only future cash flows that can be predicted in this way, but also the margins to be deposited or various risk and profit/loss KPIs (e.g. cash flow at risk (CFaR) or sales revenue) In this respect one of the key tasks for the setup of predictive analytics methods is defining the data that is relevant for the calculation of individual performance ratios.
Many companies are already aware of the enormous potential. The current trend is shifting from decentralised planning to the automation of self-learning projection models and to growing investment in the use of internal as well as external data sources to improve forecasting.
The short term optimisation of intraday load forecasting for the avoidance of balancing energy or the resulting quarter-hour trading cannot be realised without the application of digital technologies. The use of information available at short notice to determine adjustments to the day-ahead forecast requires speedy integration of data from various sources as well as self-learning pattern recognition algorithms. An automated trading system is indispensable for immediate implementation across many small-scale purchase and sales orders. Demand for such solutions from energy trading units is driven by competition with market participants who already make use of this potential, and the perpetual pressure for efficiency, and it leads to an increasing offering of such solutions by system manufacturers.
But where should you start with digitalisation and how should trailblazers in digitalisation carry on? The large and ever-growing number of buzzwords on the topic can often mean overview gets lost along with the focus on what's important; for this reason it is important to assess individual application cases and technologies at the outset. Doing so requires comparing the current degree of maturity of individual technologies with the potential benefit in your energy and commodities trading – from this it quickly becomes clear which application cases to ignore (low benefit and low degree of maturity), observe (high benefit and low degree of maturity) or implement (high benefit and high degree of maturity).
In this context an assessment should be carried out to detect any opportunities already offered by the current system landscape that are not being fully utilised. There should also be an assessment of the extent to which the further development of methods and processes is a necessary requirement for the implementation of digital solutions.
For a medium-sized public utility this could mean:
The matrix of all application possibilities and technologies resulting from the work step should form the basis for strategic discussion on the roadmap to energy and commodities trading 4.0 and it should be updated continuously.
The digital transformation is in full swing. The forerunners in the financial sector are evidence that it pays, both strategically and economically, to seek out sensible potential applications and to be rigorous when implementing them.
Companies in the energy and commodities industry are catching up. Many companies are already investigating and implementing established technologies, for example the increasingly mature algorithmic trading systems for the intraday market, or cost-efficient sensors for evaluating maintenance requirements at power stations and wind farms (predictive maintenance). In addition to revenue and efficiency, they report improved customer retention and the opportunity to exploit new markets. The promising technologies that keep resurfacing illustrate that we are currently standing at the beginning of digitalisation and that we have an exciting journey ahead.