What is Big Data? Analytics, definition, meaning & examples
Added value of Big Data in logistics
A brief glance at possible application scenarios shows why intralogistics in particular should concern itself with the value of data. "An analysis based on historical and real-time data can help to illustrate the complex, highly variable logistical interrelationships in a simple and transparent way and serve as a basis for decision-making. In this way, decision making processes for warehouse technology are also accelerated and made more reliable," outlines Prof. Michael Schenk, Director of the Fraunhofer Institute for Factory Operation and Automation. "The automated capturing of the operational status of the warehouse is also a permanent source of data that allows novel, modern analysis methods to be used. In dynamic day-to-day business, they can ultimately be used to make even more reliable decisions and increase warehouse availability while simultaneously reducing inventory levels and resource consumption." One of the results of this development could be that, for example, innovative software will allow inventory management to be fully automated in the future. However, Fischer points out that this will not happen overnight.
Examples for Big Data applications in logistics:
- Utilising traffic forecasts: CEP service providers and other transport companies have to deliver ever faster in order to prevail against the competition and meet customer requirements. Real-time traffic analysis is a blessing for them - and Big Data makes it possible: The storage of movement data from countless road users allows precise forecasts to be made, e.g. where congestion can occur and how long alternative routes will take.
- Revolutionising risk and inventory management: Production loss due to a shortage of raw materials is a horror scenario for manufacturing companies. Thanks to Big Data, inventories can now be managed much more efficiently with the help of appropriate software. The analysis of consumption data allows to generate warning systems that notify the company of any bottlenecks in advance.
According to experts, Prescriptive Analytics is still far too rarely discussed. It is no longer just a matter of giving recommendations for action, but of determining which is the best option available. This is based on mathematical methods that can suggest different solutions based on Big Data and also show how much these options deviate from the optimum. "We shouldn't be too fixated on the idea that at some point machines decide everything. However, the proposals for decisions can be made by the machine and the person can make the final decision. Once we have built up enough trust, there may well be automated solutions," says Dr. Fischer. "It will be precisely this data that will be used to create new business models in intralogistics. The prognosis is like a weather forecast. But through Prescriptive Analytics, we will be able to influence and change the weather."
This insights will provide a further incentive for companies to become involved sooner rather than later with software for capturing, processing, and evaluating Big Data: Other - possibly entirely new - players will do so. Online giants like Amazon and Google are already exploring markets for new ways to leverage their expertise and skills in new industries, and start-ups are springing up in a variety of industries - some with developments that established companies haven't even thought about. Suppliers of steel and iron products, and of services that are primarily tailored to specific industries, are still relatively unaffected by this development. But, just as intelligent IT solutions are becoming increasingly important to customers, it is only a matter of time before this protective wall collapses.
Where does the data come from?
There are sufficient data sources for Big Data in intralogistics and there is already a large amount of data available. You only have to think about how much data is generated by forklift trucks - movement sequences, routes, goods codes, and so on. A large part of this data can already be collected via control systems, but they are rarely used because the necessary tools are not known and trust in Big Data is still low.
However, if one looks at the notion of the Internet of Things (or cyberphysical systems), it becomes clear that innovative software is urgently needed to handle Big Data. "We are still at the beginning of this development," says Schenk. "On the one hand, the required technologies are available. Thanks to the possibility to network mobile objects with new radio standards, we are even advancing to a new level of technology. In addition, the prices for the technology are dropping while at the same time their performance is improving. On the other hand, there is a lack of interface standards from source to sink. Logistics urgently requires an international and standardised availability of telecommunications infrastructures. This is a priority that must be addressed in the coming years."
But once these steps have been taken, companies face a flood of data that can no longer be traded conventionally. Only machine systems can recognise the patterns that can then be used to generate added value. In this environment, this will also happen decentrally, with objects communicating directly with each other, finding the best local solutions for current problems, and then integrating them into an overall context. The accumulated data can be used in the sense of Big Data to gain new insights.
In this short video Great Learning presents five exciting trends on the subject of Big Data.