With emerging technologies like IoT and big data, we witness a revolution; ‘Fourth Industrial Revolution’
Isn’t it good to witness such a big thing? So let’s talk about those avant-garde ‘things’ but, with a big addition; Platforms.
A platform is like an operating system for a PC; you cannot manage your hard drive or graphics card without an OS, nor you can deal with big data and IoT without a Platform. We will deal those ‘things’ with a specially tailored platform for IoT like KocDigital’s Platform360.
First of all, let me tell you about the transition process to Industry 4.
When Richard Morley finds PLCs (Programable Logic Controller) at 1969, this helped to grow in the production capacity, speed and quality. And these factors pioneered 3rd industrial revolution. And now with Industrial IoT, Big Data and Platforms; production capacity, speed and quality goes to next level.
So now; we know the revolutionary factors, but what changed and how does it affects the Industry 4.0? Factories use PLC among with machines/robots etc. and communication has already been conducted via PLCs. But how about other machines or humans or environment? Unfortunately no! In the context of digital transformation, someone who looks from the outside may have the impression that all robots will change or robots will replace humans. But what is actually done is to obtain data by increasing the communication of existing robots and making use of this data for different models to actually decide on which actions to take for increased efficiency with platforms.
Let’s start with IIoT. Industrial IoT is a communication network which uses variation of communication technologies like wifi, bluetooth, rfid or zigbee. It’s track, gather, analyze and modify information. And with that, industrial companies creates systems which helps to make more efficient decisions and creates valuable information. An IIoT system contains below elements;
- Smart entities that can detect, communicate and store information about themselves (i.e. applications, controllers, sensors and security components)
- Data communication infrastructure (Can be cloud)
- Analytics and applications that generate business information from raw data
In fact, the biggest benefit of IOT is that a factory can now easily receive data from many sensors and process it instantly. And this enables applications like predictive maintenance, asset tracking, better field service, facility management etc. and of course Analytics.
But wait, there is a problem; we have a lot of data. Big Data
Often, big data refers to larger and more complex information sets than companies can use. More advanced technologies are required for such data that traditional software cannot handle. That is data that comes in increasing size (Volume), velocity (Velocity) and contains many different information (Variation). Also known as 3V.
- Volume: Collects data from a variety of sources, including organizations, transactions, social media, and sensor or machine-to-machine data. Although it was difficult to store data of this size in the past, it is now possible thanks to developing technology.
- Velocity: It is the rate at which data is received and transferred. Normally, the highest throughput is achieved by transferring directly to memory rather than writing to disk. Some smart products work with real or near real time data streams.
- Variety: Data can be in any format, from structured, numeric data in traditional databases to unstructured text documents, email, video, audio, and financial transactions.
Now with big data we have a lot of information and we can be more precise and find more answers. But the main point is not the quantity of data; it’s what we can do with this data. With analysis of this data we can;
- Reduce cost
- Better production speed
- Develop new products
- And most important, smart decision making becomes possible.
But those are hard without platforms. Somehow we got the data and store it. This data can contain xml, json or just a numeric value. How can we unify or create structure for this data and it’s source?
How Can IoT Platform Help This Revolution?
IoT platform is what separates a factory from a smart factory. At the least, it is expected to have these capabilities;
- Transports/collects data from devices and sensors
- Must be able to store large amounts of data with the least cost
- Work with large amounts of data (Not just store)
- Should be able to edge analyze
- Should be able to perform real or near real time control and analysis
You can ask, “Why a platform need to do those?” so let’s look at the manufacturing in terms of software architecture. It contains those five key elements:
- IoT Devices
- Manufacturing Applications
- Enterprise Applications
- IoT Platform
- Data Visualization and Control
Maybe a infographic can explain better. Take your time and examine the infographic below.
As you can see platform sits in the middle of the manufacturing process and orchestrates most of the process. It controls devices and collects data from them. Interacts with IT applications and enables visualization of the process.
Within this point I ‘m gonna go with real-world scenarios, cause basically we can think of the IoT platform as a facilitator to focus on business problems.
Real-world Scenarios With Platform360
As I mentioned before, at KocDigital we developed Platform360. Platform360 is a world-class IoT platform, we developed it by referencing the oneM2M Standard. OneM2M enables unified communication across OneM2M enabled devices with using our platform without effort.
With Platform360 we built solutions to some industrial areas;
- Manufacturing Execution System
- Remote Asset Monitoring
- Predictive Maintenance
- Free Floating Car Share etc.
Let’s continue with ‘floating car’ scenario. On this scenario you have a business idea and you can create an application, but you don’t have any cars. We will talk about two scenarios here; with and without platform.
For those two scenarios let’s assume those;
- We are a small team like 3 or 4 people and for market interest we have a 4 months to go live
- Within this time period, car tests takes minimum 1 month
- Car A: IoT enabled and can communicate with using oneM2M standard
- Car B and C: Our average cars, does nothing. Needs digital transformation
- IoT Device X: Car B only accepts this device’s interface
- IoT Device Y: Car C only accepts this device’s interface
- Device Manufacturer 1: Manufactures IoT Device X and they can communicate only with their own standard, but negotiable only if there is a good solution which benefits them too
- Device Manufacturer 2: Manufactures IoT Device Y and they can communicate only with oneM2M standard, not accepts negotiations about implementation
- At basic level we can be able to send commands like open the door and get car telemetries like fuel level, gps etc.
Scenario 1: World of Chaos :)
In this scenario we don’t have a platform, and somehow we need to talk with cars. We can create our own standard for machine to machine (m2m) communication but it takes too much time and effort.
The main thing is, this m2m communication, security, management etc. should not be our interest. We need to focus our business idea
For issued 4 month ‘Time to Market’ duration(TTM) we’ll gonna skip implementing standards like oneM2M or lightweightM2M and take consultancy service from Device Manufacturer 1 and Device Manufacturer 2 for covering multiple types of cars. This alone can take your TTM duration and budget. Somehow you can talk with Car B and C. And for Car A, Device Manufacturer 2 gives you a option; they can help you to talk Car A but with a price of course. They will gonna talk with car using oneM2M standard but they won’t talk about this to you; they are gonna tell that, they’ll attach IoT Device Y to Car A.
Now let’s thing about the budget killer mess for just communicating.
- Consultancy prices
- IoT Device X and Y prices for car B and C
- Extra price for Car A cause IoT Device X
- This negotiation procedure takes minimum one month. So add operating expenses for a month
- Within this negotiation process you can start to develop your business idea. But due to new finalized data models most likely your application gonna need a refactoring. Add this price too
- Communication infrastructure price. You need to get a database cluster for IoT data and maybe queue. Include license price too
So basically, you can start developing your business idea after the first month with a narrowed budget but you only have 2 months left. And most likely, you need to take note which car uses which standard or request response model. And again, most likely there will be flaws within implementation. I’m not event talking about performance and security you don’t have time to think about those. And with a high chance, stored IoT data model won’t compatible for big data or AI applications.
What happens if there is a new car which could not use neither IoT Device X nor IoT Device Y ? You need to welcome new consultants and budget holes :)
As you can see, it’s nearly impossible to succeed in the given TTM duration. It takes lot of risks, uncertain points and creates unnecessary budget holes.
Scenario 2: Power of the Platforms
Let’s use a good IoT platform, like Kocdigital’s Platform360.
But first what is the key benefits and capabilities of the Platform360?
- SDK: There is a Device and Service SDK, they supports multiple languages like .net core, nodejs etc.
- Native OneM2M Support: Effortless connection with devices which supports oneM2M standard.
- Performance: Developed using with state of the art techniques and technologies. It’s stateless and can be easily scaled.
- Security: We comply with current valid security checks.
- PaaS, SaaS and On-premise: Supports multiple models with competitive pricing.
- Easy to use: With well defined guides and documentations it’s easy to use.
So we read the docs and at day one, we know our data model, request and response types and what to do for communicate. We can start to create our business idea without losing time.
Car A and Car C ( with using Device Manufacturer 2’s IoT Device Y) can communicate with us using oneM2M standard. This will not take much effort.
Car B is a bit problematic but we have options;
- We convince them to use oneM2M standard. It can be good for them too.
- We or they can write a IPE (Interworking Proxy Entity) for communicating
- They can create a application layer using Platform360’s SDKs
Either way most of our options probably takes time and money but the point is we changed noting at our business or architecture. Everything stays still and this will be like for new cars which could not use oneM2M or IoT Device Y.
Let’s examine our expenses;
- Platform price. There are multiple pricing options, we can start with a minimal configuration. Includes support, infrastructure etc.
- Consultancy for Car B. This is nothing big like scenario 1.
- Operating expenses. But mind that we start our business at day 1
- Nothing :)
Isn’t it just great? There is no turmoil, no budget holes. Just focus to your business and create value for world and yourself!
Thank you for listening to me up to this point. I try to explain the whole process and give real-like examples.
When we compare the two scenarios, we saw without platforms digital transformation is hard and needs too much things. And it’s prone to being unusable.
It’s safe to say; a smart IoT solution needs a platform with a natural way.