If you think that big data projects are challenging, then you probably have not seen yet what the Internet of Things (IoT) projects entail. It depends not only on size, but also speed. IoT consists of getting the Big data which can beunstructured, furthermore, this real-time data needs to be processed and analyzed. The speed at which this is done is enormously important.
Therefore, often the size of big data projects is just a piece of cake compared to the size (and speed) of IoT projects. Twitter is a good example, because it is sometimes referred to as a source of big data. Through the website Internet Live Stats, we see that the number of tweets per second is estimated at between 6,000 and 8,500.
The big difference between IoT and big data projects (lack of) time, so that short period of time is important. In big data projects, it is perfectly normal to rest the data before they are used for analysis, while IoT projects, time is essential;The analysis should take place as soon as possible after the data is received.
That’s whyIoT projects areFAST data projects.
To make matters even more complicated, there is still some considerations when it comes to speed. The first problem is that data from mapped device often are delivered in simple format. A format that is suitable for decision without further processing. That data must therefore be organized, transformed and enriched to be valuable for analysis.
ORGANIZE AND TRANSFORM
The unorganized in coming data cannot be analyzed directly. The efficient re-arranging of in-flight data is a technological challenge.
Transformation is necessary because the original data can be used, but it is often not sufficient for advanced analysis. This may require some more explanation. Let’s say you want to determine whether a value is within the range of what is considered normal, or that it is an exception. In a simple analysis you could do a static analysis, where hard coding your lower and upper limits. In practice this would probably cause a lot of false positives, the reliability of the system is not beneficial.
It would be better to be able to do this dynamically by performing in real time a Bollinger band-calculation. The Bollinger band may need an Exponential Time Based Weighed Moving Average (EWMA). This EWMA may be regarded as a derived value, which is to be calculated every time there is a new event. This is an example of a real-time transformation.
The third and final change concerns the enrichment. Which is necessary if the decision models not only use data from the devices, but also other resources needed in the organization. These could have agreed on the service level agreement (SLA) when it has been committed for the last time maintenance and the historical reference data can be used.
The implications of theconstant retrieving of this data to enrich & analysis must be carefully consideredon the existing IT infrastructure. In order to ensure that the performance of such systems, therefore does not decrease in such a way, additional measures are necessary.
In order to support these three real-time capabilities has advanced a business integration, analysis and in-memory caching capabilities required.
Big data was the first step; what the Internet of Things needs is big FAST data.
THE FIVE STEPS OF DIGITAL TRANSFORMATION
Digital transformation is hot. Most organizations know they need to make the change. But how? With these five steps, you are well on your way.
An organization is essentially the sum total of its physical, financial, human, intellectual and relational abilities. Manufacturers invest most of their capital in physical assets. High-tech companies invest mainly in R & D to create new intellectual capital.
But with the changing technological landscape, where you must particularly invest in?
Digital transformation: leaders and followers
In today’s market we see that especially the relational capital is growing enormously in technology platforms at almost zero cost. Think of Facebook, LinkedIn, Match.com, Uber and Airbnb. Even if these companies rely on physical assets such as cars for Uber, they mostly have the technology, not the physical assets.
Meanwhile, those left behind continue to spend their time and money on assets that are not so easy to scale: physical goods (such as factories or inventory) and human capital (such as highly trained staff providing services).
Digital transformation requires companies to reallocate their capital on new business models that support the digital transformation.
Especially leaders who feel comfortable struggling with older types of assets to digital transformation.
How to manage such a lengthy and uncertain process?
Although digital transformation in every organization will look like a different, we can distinguish five steps.
Step 1. What is your position?
The first step is to determine establish your starting point. That includes identifying the current mix of assets and the business model that creates your asset portfolio.
For example, your company makes and sells stuff? You hire skilled employees and delivers services? Do you develop software or medicines? Build and manage digital networks?
What habits and patterns block the transition to a digital organization?
Whatever it is, a clear view of the point of departure gives insight into strengths and weaknesses. This allows you to discover the longstanding habits that need to be changed to create a digital transformation.
Step 2. Make a complete inventory of all the assets of your organization
Start with the simple things that you have always followed. The physical assets such as plant, property and equipment.
Followed by the lesser-known intangible assets such as the talents and skills of your employees, the networks of people and organizations that exist outside the traditional boundaries of your business. These assets are often overlooked, undervalued and inadequately managed.
Look especially at your networks to determine their size and affinity. Work with a network involved.
Step 3. Visualize the digital future
Visualize a new future in which you co-create a digital network with partners in your external network. Customers and suppliers are often the first choice, but there are many more to consider networks. Consider alumni, distributors, integrators, investors, communities, and even competitors. The goal is to find a way in which you and your network would create value together. Using both your assets and theirs.
This is a difficult step, but keep in mind the four types of assets that you and your network need to create value together.
- You could give access to physical assets such as Airbnb and Uber.
- Generate new content, such as Yelp and TripAdvisor.
- Offer services such as Task Rabbit and Mastercard.
- Generate networks such as Facebook and LinkedIn.
Consider also the digital platform to facilitate this cooperation.
But start small, with a small investment and learn from each step.
Step 4. Start with a pilot
Start with your business network in a pilot. Move above a small portion of your capital (including time, talent and money) to the new initiative. This requires that you build a team. Fund this team and give it plenty of room to learn and change, without too much bureaucratic oversight.
Put a team at work, give them space to experiment
Spend a lot of time (in this plan!) To communicate with your audience network to ensure that you are creating a platform that they want to contribute and they find worthwhile.
Step 5. Follow the progress of new KPIs
Keep track of your network initiative closely watched. This requires reporting on new statistics. Probabilities Truly you have to set new key performance indicators (KPIs) for the digital network model.
But first, work on new leadership
Organizational transformation must begin with a leadership transformation. So make sure your board, CEO, and the rest of the leadership team together fully supports and is ready for the new priorities. And change the investment mix. If not, then the digital transformation does not come out of the ground, or at an early stage to a halt.