DEPLOYING MACHINE LEARNING

The Internet Of Things is enabling a transformation that makes it possible to gather and analyze data across machines, enabling faster, more flexible, and more efficient processes to execute at higher-quality at reduced costs. But how do you start? What are the steps you have to take to go from data collection to improving your processes? Let’s describe a real case we implemented to answer these questions.

We believe complexity should be hidden from the users, we believe you can apply machine learning without being a data scientists, we believe you can collect data without being a data engineer and finally we believe you can create your own dashboards, actions and control your processes without being a programmer. We do this by putting you in control of the tools.

Let’s present you with a real use case we implemented. It is a simple use case, a customer asked us if we could help monitor thousands of remote devices that are pressurized. These devices have the capability of sending their current status every hour. The question is which ones are leaking and need attention by a field service engineer.

 

We started with collecting the information, like many sensors and machines nowadays the customer can already configure the pressurized device to forward the data securely to the MEUNGO application. Actually, this customer started their Internet of Things project enhancing their own hardware to communicate the status, while doing this they used the MEUNGO software to test and verify the communication. Out of the box the MEUNGO software allows users to add and change data sources (sensors), how they communicate and how to integrate these data sources into one device type. We believe we live in a changing world and therefor allow the user to change the definition of the world, we do not program it, you as a user configure the system.

After connecting the devices and verifying the data using user defined dashboards the customer was ready to create some information out of the data. The challenge here is the amount of devices, not always knowing the desired pressure and the fact that some of these devices are exposed to outside temperature that influence the pressure. Simply detecting a lower pressure is not sufficient because that could be because of lower outside temperature, manually checking for a trend in one the user defined plots does not scale well and is error-prone.

The solution the customer selected was to invoke the simplest machine learning algorithm one can execute on a time series. The customer decided to configure the system to train a linear trend algorithm every day using a week of history. The trend of the estimated line indicates if the pressure over the course of a week is increasing (larger than zero), is stable (zero) or is leaking (less than zero). Next, the user defined a rule to be triggered only if the trend is less than zero, the actions associated with the rule are to record the event and create a work order in their field service software.

The customer was able to go from collecting data to action in less than a day and continuous adding actions based on progressive insight. On top of that the solution allowed the customer to improve their Service Level Agreement by guaranteeing the presence of pressurized devices at a lower cost.


Overall Equipment Effectiveness and the Internet Of Things

We strongly believe that the Internet Of Things will allow businesses to improve and automate their processes all the way up to autonomous execution.  In manufacturing one of the crucial components of the Total Productive Maintenance system is Autonomous maintenance. How does the Internet Of Things help the implementation of Total Productive Maintenance? How does the Internet Of Things contribute to the Overall Equipment Effectiveness? And how does predictive maintenance as something that the Internet Of Things enables have anything to do with Total Productive Maintenance?

Let’s go back to the beginning, Overall equipment effectiveness or OEE is a measure of how well a production unit works compared to its full potential. Overall equipment effectiveness is a term coined by Seiichi Nakajima in 1982 in his book TPM tenkaiHe was one of the founders of the Total Productive Maintenance system which main objective is to increase Overall Equipment Effectiveness by applying the concept of the 5S and promoting 8 supporting activities.

SEIRI

sort

The first S aims to eliminate anything that is not truly needed in the work area as for example unnecessary materials and equipment.

SEITON

systematize

The objective of the second S is to organize the remaining items in order to define and maintain clean locations for tools, machines and materials.

SEISO

sweep

The third S refers to the implementation of regular cleaning practices by dividing the manufacturing floor into different cleaning areas and assigning a responsible team to each area.

SEIKETSU

standardize

The fourth S is focusing on creating standards for performing the previous three activities.

SHITSUKE

self discipline

The objective of the fifth S is to make the system sustainable by ensuring that all standards are regularly applied.

The main objective of Total Productive Maintenance is to increase Overall Equipment Effectiveness which has three factors that are multiplied to give one measure:

  • Performance, of the theoretical speed at what speed did the production unit run
  • Availability, of the theoretical time the machine can be used how much was the production unit running
  • Quality, of the theoretical sale-able products the production unit could produce how many did it produce

PERFORMANCE

The actual speed of a machine compared to the maximum speed determines the performance

AVAILABILITY

Breakdowns result in unplanned downtime and changeover is the time between moving from one unit to another unit, it includes re-configuring a machine and the time to get a ready for production.

QUALITY

The units produced minus the defective units determines the quality.

Overall equipment effectiveness is measured as a percentage between 0 and 100% and can be reported on a machine-by-machine, product-by-product or shift-by-shifts as long as you make sure you collect enough data to be statistically significant. The eight pillars of Total Productive Maintenance are mainly focused on proactive and preventative techniques for improving equipment reliability. The pillars help to improve all the three factors Performance, Availability and Quality of Overall equipment effectiveness.

AUTONOMOUS
MAINTENANCE

Places responsibility for routine maintenance, such as cleaning, lubricating, and inspection, in the hands of operators.

FOCUSED
IMPROVEMENT

Have small groups of employees work together proactively to achieve regular, incremental improvements in equipment operation.

PLANNED
MAINTENANCE

Schedules maintenance tasks based on predicted and measured failure rates.

QUALITY
MANAGEMENT

Design error detection and prevention into production processes. Apply Root Cause Analysis to eliminate recurring sources of quality defects.

EARLY EQUIPMENT
MANAGEMENT

Directs practical knowledge and understanding of manufacturing equipment towards improving the design of new equipment.

EDUCATION
AND TRAINING

Fill in knowledge gaps necessary to achieve goals. Applies to operators, maintenance personnel and managers.

SAFETY HEALTH
ENVIRONMENT

Maintain a safe and healthy working environment.

ADMINISTRATIVE
AND OFFICE TPM

Apply Total Productive Maintenance techniques to administrative functions.

Let’s have a closer look on each of the components of Overall Equipment Effectiveness and how it is related to the Internet Of Things:

Performance

The actual speed of a production unit compared to the maximum speed determines the performance.

Improving speed starts with measuring speed and the factors that could influence speed. In a production line, speed of a single production unit is limited by the production units downstream and upstream. First thing to do is to determine unambiguously the reason of the speed, is the bottleneck the downstream production unit, the machine itself or the upstream production unit driving the speed at any given time.

Availability

Breakdowns result in unplanned downtime. Post breakdown analysis allows you to find root causes to plan downtime to prevent breakdowns by executing preventive maintenance.

The Internet Of Thing allows you to monitor your production unit as it is running. Using the data you can do several things to reduce the risk of unplanned downtime:

  • monitor actual run time of the production unit and its parts, instead of a calendar driven maintenance you apply a condition based maintenance schedule
  • detect anomalies in behavior of the production units, temperature, sound, humidity and vibration are all indicators which potentially can indicate a potential problem
  • once you have collected enough data of your production unit you can have data scientists try to discover if they can predict when your production unit needs maintenance, when does a vibration indicate a problem etc
  • detect incorrect use of the production unit, incorrect configurations both in the control parameters as in the hardware configuration

Changeover is the time between moving from one unit to another unit, it includes re-configuring a machine and the time to get a ready for production. Changeover is typically reduced by better scheduling of the production (larger batch sizes, saw-tooth changeover times etc) and by offline improvements to make it easier and faster to make the adjustments. The Internet Of Things allows you to unambiguously measure changeover times.

Quality

The units produced minus the defective units determines the quality.

Improving quality is typically done by cold analysis, so it starts with monitoring and after the data has been collected try to find correlations. The correlations will identify opportunities for you to improve:

  • Is there a relation between defective reasons and shifts, this exposes opportunities to improve by training
  • Is there a relation between type of unit and defective reason, this exposes opportunities to improve in configuration of the production unit and training
  • Is there a trend in the quality and time of production, this exposes fatigue and opportunities to improve by changes schedule

The Internet Of Things allows allows you to real time monitor the number of defective units which allows you to react immediately.

How to start with the Internet Of Things

When dealing with the Internet Of Things security is extremely important. So the first thing you need to cover is to convince yourself your risks are sufficiently covered, production can and may not be interrupted due to measuring. We understand the choices and the consequences.

Starting with the Internet Of Things can be as easy as using existing programmable logic controllers (PLC) and connect them to the internet using one of the many available gateways. An alternative to using your PLC’s is to use external sensors and attach them to your production units. Both scenarios are relatively straightforward to execute and the starting point of collecting, monitoring and reacting in real-time to your reality.

Our experience is that once the hardware problems is covered you can use the data in existing Total Productive Maintenance system immediately. We make it safe to connect, collect, visualize and react.


3 common pitfalls to avoid in IoT projects

3 common pitfalls to avoid in IoT projects

Any disruptive technology comes with the opportunity to learn fast and move forward. However, in recent months we are observing common recurring patterns. We aim to share these lessons learned to help others succeed earlier and harvest the benefits of real-time insights and actions. We also created a graphic to help describe the idea. Where are you in this picture today and where do you need to be?

1. Architecture not ready for a changing world

We hear often that change is the only constant. This has never been so applicable as in the world of the Internet of Things and accelerating technology advances. When the basic architecture is founded and rooted in the knowledge that the world is changing, you can deliver results; as only more things will be going online and you will be future proof for an ever growing number of data driven applications.

In addition to this, there will be many future use cases possible from the data that other industry partners will be interested in accessing to enable smarter decision making. Controlled access to this aggregated information (enabled by the yellow sun shape in the graphic) will have a premium value. If we refer to the graphic, in an ideal world manufacturers would focus on the infrastructure to enable and support application development by integrators, service companies and user groups.

2. Not knowing your “role” today and where you need to be, avoiding the red quadrants

Being all things to all people can be exhausting and counterproductive. In this regard, knowing your role in the new IoT world and where you should be heading are vital. We have observed the challenges here ourselves and see other industry players struggling. For example if you are an end user or service company, then the focus should not be on sensors and infrastructure development. When after-sales service and maintenance companies can harvest aggregated data, then the end-users will see enhanced service levels and up time.

3. Not realising that being a partner is as important as having a partner

In the connected world being a partner is as important as selecting a partner. Knowing your role and your partner’s role allows you to make the most out of these new relations in a very dynamic and evolving market. Take time to align on the most important issues of the goals, security and how to handle the changing world scenarios.

Fortunately there are many options available today with platforms, partners and players.

Want to avoid more pitfalls or to discuss getting back into the right quadrant,  feel free to contact us.

 


The future of the Internet Of Things

The Internet Of Things is a means to an end, and the more people we can invite into that, the easier we can make it to get to that end.

The real meat of the Internet Of Things is not the coding, it isn’t the connecting, it isn’t the dashboards. At the end of the day it’s what you’re building and the effect it has on the persons you are building it for. The Internet Of Thing is an amazing act, it’s challenging and complex, but in some ways, it can be a barrier. And what we’re interested in is the future being less about devices, connecting and coding. We think the future of the Internet Of Things is not about technology at all, we think low-code Internet Of Things is the future.

Low-code is a way to design and develop software fast and with minimal coding. It enables skilled people to deliver value more quickly and more reliably. Low-code is about getting more done in less time. Low-code, is a trend in general, it is about connecting the dots without creating the dots. We believe the Internet Of Things requires to be its own dot in the low-code community.

Low-code Internet Of Things

The Internet Of Things can be complex and requires a lot of things to be right. It involves sensors, embedded software, communication protocols, back-ends listening to devices, security, integrations with external systems, analytics, front ends for configuration, verification and communication of results and so on. One of the reasons there seems to be so many IOT platforms is because they all serve, rightfully so, a part of the puzzle that needs to be solved.

We believe that low-code Internet Of Things will be the tool that will allow you, as an owner of devices, create value for your customers. We at MEUNGO provide a low code device management software as a service which allows you to manage your own devices, users, dashboards and actions.


How to visualize your Sigfox devices?

Track assets where they are and what they do using a smart Sigfox sensor

Monitor water usage using a smart LORA flow meter

Combine Sigfox, LORA and many more sensors into a single dashboard

In this blog we explain what is required to connect and aggregate data of your devices and start creating visualizations. At the end we describe how you can start connecting your devices using the MEUNGO platform.

Sigfox is a communication provider which provides global, simple, cost-effective, and energy-efficient solutions to power the Internet of Things. Already today, there are many different sensors which use Sigfox for communication. There are sensors for monitoring buildings, environmental conditions, temperature, smart tracking of assets, buttons, smart parking and many more.

MEUNGO allows you to aggregate all your devices in one application, quickly build dashboards, build rules that listen and react to the data as it is coming in. You can manage your users and who can edit and view your dashboards.

Requirements

  • One or more Sigfox devices with access to the Sigfox back-end. The device manufacturer, re-seller or Sigfox is the one that can provide you with access to this back-end.
  • Access to the MEUNGO platform

Once you purchased your device and are able to find your devices in the Sigfox back-end you are ready to go to the next steps. For any data source (technically we consider the Sigfox device a data source) you will have to tell the platform:

  • What data is collected, attributes such as temperature, humidity and location
  • Where the data is coming from, the connectivity in this case Sigfox
  • How the messages should be translated, how to translate the incoming payload. Easy to use converters are embedded in the platform
The general page of the data source definition

Once you have finished these three steps you are ready to define a device type in the MEUNGO platform. A device type will aggregate one or more data sources together. If you have for example a machine composed of many sensors you can model it as one device type.

The general page of the device type definition

Here you will define:

  • What data you want to associate with this device type and if it comes from a data source or if the user sets it
  • Rules you would like to be applied, this can be simple logical expression or complex algorithms triggering alerts, e-mails or other IT-systems
  • How long you will store the data, in this case 52 weeks
  • Security settings, what certificate to use if third party applications would like to have direct secure access to your data

 

Definition of temperature for the device, the temperature is sourced from the temperature attribute of the data source

Next you have to tell your Sigfox back end to forward the data to MEUNGO. For this you will have to login to the Sigfox back end and define a callback function. This callback function will forward the data to an address provided by MEUNGO.

Now you are ready to connect, collect, react and create dashboard for the devices you have.

Dashboards

A dashboard is an interactive visual representation of the collected data, it can be a snapshot of the latest received information or it can be dynamic and update while information is coming in. The MEUNGO platform allows you to manage organisations, users, devices, data, rules and dashboards:

  • You can create dashboards using any combination of devices your organisation has access to
  • You can share dashboards with any subset of users in your organisation
  • You can look at multiple different type of devices at the same time in one or more visual components, this allows for easy comparison and visual verification

Connecting, collecting and visualizing your devices is the starting point to become a high-performance data driven organisation.

You can start today connecting your devices.

Evaluation licences to the MEUNGO platform are available, contact us and we can get you up and running on the same day.


Connecting multiple different Sigfox devices into one application

Sigfox is a  cellular style system that enables remote devices to connect. It is a technology which lets your devices communicate just like 3G or 4G on your phone. It allows you to send 140 messages per device per day, payload size is 12 bytes and throughput up to 100 bits per second, has a very low energy consumption and has excellent coverage around the world. All in all a good communication solution for many use cases. There are many devices/sensors on the market which communicate using Sigfox. We have used several of these Sigfox devices for customers and have also tested a few of them, see for example the posts on 1608 and Sense'it.

How easy is it to integrate a collection of different devices into one application?

To answer this question let's have a look what needs to be done:

  1. You need to tell your application what kind of information it will collect and how it will be collected
    • What sensors are being used, how do they communicate and how do you translate the information from the sensor so the application understands it
    • What sources of information do you invoke when, if you want to reason about the weather when does the application ask for the weather and who does it ask
  2. You need to tell your application what you want to reason about and collect information for
    • If you are a machine builder, re-seller of equipment, rent out equipment or a fleet owner you might want to reason about machines or devices
    • If you are a facility manager you might want to reason about a room, a building or the people in the building
    • If you are a farmer you might want to reason about your crop, stock or farming equipment
  3. You need to connect the collected information with the things you want to reason about
    • If you reason about a room in a building with three different sensors and using the outside weather you need to organize your application such that all these four information sources are collected for a single room. Once this is done you can easily organize your dashboards, deploy your actions and execute your analytics.

Simply stated, what things are we going to reason about, what sensors are my things using and how do the sensors connect to my things.

In the MEUNGO platform you model the sensors (node types) that collect information by defining what they collect (attributes). Next you define how it is communicated (connectivity) and finally how you translate the information that is communicated to an attribute (conversion). In this blog we show how to do this for Sigfox. The platform has a secure connector to the Sigfox back-end and will monitor the sensors once they have been provisioned. Other connectors for other communication protocols also exist.

 

In the MEUNGO platform you model your things (device types) by defining what attribute types they have and if the information comes from a sensor and if so what attribute of that sensor. So if you model a room where you are going to collect outside temperature and inside temperature you make two attribute types, one matched with the sensor collecting inside temperature and one matched with the sensor of the outside temperature.

 

Once you have defined your device types you create instances of them (device instances) and connect the sensors to the instances by providing identifiers.

 

Once this is done your data will be stored and is accessible for dashboards, actions and analytics.

Real world applications have to deal with a changing world. New devices will enter your world, new versions of your current devices will be released. Over time you will have to maintain a hybrid collection of devices in the field. Your technology stack will have to be able to deal with this fact of a changing world.

MEUNGO is designed to easily connect a collection of sensors into one object, device or thing. MEUNGO makes it easy to connect, collect and react and allows you to define and share dashboards. It is not limited to Sigfox and supports other communication protocols.

If you have one or more Sigfox devices and you want to start drag and dropping your own application feel free to contact and see what is possible.

For a list of devices using Sigfox see https://partners.sigfox.com/products/device

 


Robotics and IOT: Predictive maintenance, why?

When most people think of robots, they either envision humanoid servants or big one armed machines on automotive production lines. At MEUNGO we think about robots as being autonomous systems, things that happen by themselves. One of these things that could happen by itself is planning and dispatching preventive maintenance activities. Preventive maintenance prevents unscheduled downtime, it prevents having to stop production, call for service and potentially have unhappy customers.

So how do you prevent unscheduled downtime? The obvious answer is to plan downtime before the machine requires immediate service to continue operating. For this we need to know what the conditions are such that the machine no longer can operate and when these conditions will happen. So if we know when the machine is going to breakdown we know when we have to plan scheduled downtime. The rather dramatic scientific name being used for determining when the machine is going to breakdown is determining the Remaining Useful Life, the RUL of a machine. Determining the RUL helps you in predicting when you need maintenance and we do this to prevent unscheduled downtime. Predicting maintenance is input for your planning and scheduling tool to schedule downtime at a convenient time for you and your customers.

There are many methodologies to determine the remaining useful life of a machine. Almost all of them have one thing in common, they need data as input. Data about previous breakdowns of similar machines, parts, environment variables etc to use in some mathematical model to predict the RUL. Here one get can get pretty fancy pretty quick and use artificial intelligence, filtering techniques and stochastic processes to take into account as much as possible. One thing is critical, these advanced techniques typically need a lot of clean reliable data.

A more pragmatic and definitely a good intermediate step before invoking sophisticated RUL methodologies is to predict your maintenance using the condition of your machine. Let's use your car as an example, you observe how many miles you have driven but you also observe if your steering wheel is shaking. The earlier you detect the shaking the earlier you can schedule a maintenance to check your wheels, brake rotors and brake caliper. So instead of predicting when your machine breaks down you act as soon as you observe an anomaly or reach a certain condition such as mileage or a date. You prevent unscheduled downtime but probably are sub-optimal because your RUL could have been months. Once again what is critical here is that this methodology also needs data and also continuous monitoring.

At MEUNGO we think you need to start collecting data now independent of what fancy technology you are going to use to predict the remaining useful life of your machines. Also start robotizing now, make sure conditions and conclusions on your data are translated into actions. For example make sure when your data tells you there is a problem your field service software auto-magically gets a work order. The trigger to when to auto-magically create a work order might change but the infrastructure to do so does not have to change. Once you do this you have the basis building blocks in place to continue improving your predictive maintenance.

 

 

 

 

 

 


Robotics IOT in Retail: Vision based inventory and shelf condition

In my previous post I talked about robotics and IOT. In this post I will give an example of such a robotized system in retail.

One of the trends in grocery retailers is transforming their stores to strengthen their high quality fresh offer, either for take-away or for consumption on-site. For example sushi and healthy meals presented at a premium spot in the store.

Fresh products have short shelf life and need a climate controlled environment. As a retailer you want a high availability of your product and prevent out of stock which results in unhappy customers. The combination of high availability and short shelf life creates a high probability of shrink, having to throw away products past expiration date. A strategy for preventing this kind of shrink is to have some markdown policy on items that are about to expire.

Besides the availability and price the presentation of these high margin products is critical for the success and experience. Current strategy is continuous monitoring  by the front-line employees to maintain a good shelf condition typically by keeping the shelves fully stocked.

The Internet Of Things connects the physical world to the digital world using the internet. Sensors produce the data and they can be connected to all kinds of things. Here we can think of monitoring the climate controlled environment but also we can monitor the shelf condition using images and vision recognition. Several companies are already developing and using systems: Amazon Go is probably the most advanced, the LoweBot of Lowe's and  Bossanova tested by Walmart are all good demonstrations of using vision and robotics.

CLICK ON IMAGE TO START VIDEO

CLICK ON IMAGE TO START VIDEO

CLICK ON IMAGE TO START VIDEO

Unfortunately we do not all have such deep pockets to innovate like these big companies. But in today's open source software world you can relatively easy get a good vision recognition system up and running in a short time. This student at Stanford Robotics demonstrate in a well documented project you can easily monitor some shelves to detect out of stocks using free open source software. Learning from this project it is relatively easy to start using a fixed camera to monitor critical shelves to collect  business intelligence and to create alarms when the shelf condition is no longer what it is supposed to be. The collected data can also be used to implement a better dynamic pricing scheme focused on revenue optimization instead of spillage minimization.

If you ever felt that you would like to know more about what happens to your shelves, you want to increase your business intelligence and improve your operations then act today. The first steps are quite easy and consists out of monitoring more of your operations. If you do not know what to do, then do something and give us a call.

 

 

 


Robotics and Internet Of Things (RIOT)

I really like to make things efficient, effective and automated. My professional career has been about planning and scheduling, helping others make things run more efficient. I have created software tools to plan factories, schedule people, route trucks and planes and to make NASA robots and drones operate autonomous. In my last role as an employee I combined my knowledge in regression modeling, forecasting and optimization to price products of a large retail organisation based on billions of receipts. Today, one could say, I do something completely different. I don't see it like that but I am helping organisations to connect devices to the internet. Let me try to explain why I think that makes sense for a planning and scheduling guy to run an IOT platform.

Gromit (red robot) in interaction with K9 and an astronaut in 2003 at the NASA Ames Roverscape

In planning and scheduling we match demand and supply. Something or someone needs resources typically with some time constraints. If you know exactly what you need to do we talk about scheduling and if you also need to determine what to do we talk about planning. Sometimes you just want a solution to make sure you can execute it. In other cases you want the best solution, you want the one that is most cost effective, robust, highest profit, lowest carbon footprint or whatever makes sense for you.

In all planning and scheduling problems you need to have rules, a goal and data. Of these three things, rules and the goal typically do not change that often but data can and will change. If you schedule a timetable for your buses in town the data does not change very often, maybe every quarter or so you make a new timetable based on new demand, roads and employees. If you plan and schedule an autonomous robot like a car, which continuously gets new data from its sensors, you plan every time you get new data that makes the previous plan inefficient or impossible to execute. This might involve planning and scheduling every 10 milliseconds. In both cases you need to solve a planning and scheduling puzzle but there is one big difference. In case of making a timetable you have a bit more time to solve the puzzle, in case of a robot you might have to plan an emergency stop because the robot is going to hit a pedestrian. Therefor in robotics we typically have multiple planning algorithms at the same time, reactive planner which determine what to do next and deliberative planners to make sure we reach the goals we have. The challenge is to keep them in sync.

To do robotics, which is a system that operates automatically to accomplish a goal, you need computing power, algorithms, a model, continuous updates of data and systems you can control to execute. This is where connecting devices to the internet becomes interesting. Whenever I was working in planning and scheduling for businesses we were solving problems for tomorrow and execution was done by people. I strongly believe that today we are bringing together trends that will enable us to run business processes autonomous. For this we need at least reliable continuous data  acquisition, it is for this reason I am now connecting devices to the internet. IOT not for the sake of connecting but to create the fuel for autonomous planning and scheduling.

MEUNGO provides the tools to connect devices, monitor and act on your data with a vision to robotize your processes.

 


Sense'it

The Sense-it 3 is a fully featured piece of hardware for measuring all kinds of things. In this article I will share my experiences in using this device to start tracking assets including temperature and humidity.

The device

The Sense-it is a nice looking device equipped with a single button. It comes in a small box and can easily be brought online with the provided instructions.  It can measure temperature, humidity, light, door opening and closing, vibration and magnet but not all at the same time, you will have to select a single mode. You can configure to send data on request by pressing the button, have it be triggered by some event or every hour or period you like. The communication goes via Sigfox.

It has a nice web application where you can manage all your Sens-it devices and where you can do some more advanced configuration of your device. You can also download a Software Development Kit (SDK) which allows you to use the sensors however you want. All in all a very nice device which gives great out of the box functionality and also allows more advanced use cases.

I wanted to test how I can integrate this device into the MEUNGO platform. The MEUNGO platform is intended to integrate multiple different types of devices and act on the aggregated data. Sense-it gives you the option to define notifications to send e-mails, SMS or invoke web-hooks (callbacks) when data of the device arrives.  Although I could have used the web-hooks to forward the data from this device I decided to go one level deeper and use the web-hooks in the Sigfox backend to forward the data to MEUNGO. When you do this you can no longer use the Sense'it web application to configure the device. In this setup Sense-it is no longer connected to the device, if you really care where your data goes and don't want Sense'it to have any then this is an option.

To make the Sigfox forward the data as soon as it arrives in the Sigfox backend I configured a callback in the Sigfox backend. MEUNGO has a standard connector to the Sigfox backend so this is plug and play. Next the MEUNGO platform needs to understand the payload. Sense-it has this very well documented and together with the payload converter in the MEUNGO platform this was up and running in a heart beat.

In the post 1608 I described how to get the 1608 device online so now the Sense'it is online we can compare the sensor readings from both devices. The MEUNGO platform allows you to configure custom dashboards in which you can drag and drop all your devices in several different visualizations.

In this dashboard I added three visualizations:

  • one Map which only will show 1608 locations because the Sense'it does not have GPS
  • one Gauge which shows the real time temperature reading from both devices
  • one Graph which shows the historical humidity and temperature reading from both devices

In the left upper corner you can make a selection out of all registered devices. The N-001 device is the 1608 device and the S-001 device is the Sense'it device. It is beyond the scope of this post but it is interesting to see that, although the devices are sitting next to each other on my desk, the readings can be quite different especially humidity.

Conclusion

The Sense'it is a very nice device, it works out of the box. You can use the node as is but also completely configure or program it to your needs. The documentation has everything you need. It was no problem to connect the device to the MEUNGO platform and to integrate its data with other device data coming from other sources. A nice improvement would be for the device to have GPS plus a more technical thing to allow the user to use callbacks like Sigfox (especially cross-account AWS callbacks).