Value proposition of Telemetry for industry 4.0


In the age of precision farming, the role of technology is decisive and amplified much higher than ever. There are many ways technology can help industries to improve the offering and keep renovating. First usage of electronic system in automotive industry was vacuum tube car radio in 1930s and from then to now, manufacturing cost related to electronic components has raised approximately to 30% of the total cost of the machine. As the machine becomes more and more electronic, there is an increased scope of monitoring the machine, by sensors and signals.

Installation and tuning of sensors are massive activity for the OEM and some of the major areas of system applications of main control function(s) are (a)

As the application of sensors are increasing, data and insights generated using these sensor are becoming more predominant. Emerging technologies like IoT, Artificial Intelligence and predictive analytics are providing an opportunity to embrace this digital change. All such areas and many more are doing their bit towards Industry 4.0

Where do we stand in agriculture business?

From a longtime automotive industry has been driven by the car industry. Any new innovation in car industry will be adapted by other segments based on the applicability and is not different for agricultural industry as well. Whenever there is a breakthrough in the car industry (Digitalization, Autonomous driving, electric motor) those will be slowly adapted by agricultural industry. It wouldn’t be in incorrect, if I say Agricultural industry is the succeeding generation of car industry.

But there are lots of differences as well between these two spaces. Agricultural industry is driven by necessity over luxury, machine usage is generally seasonal, closer watch on warranty, higher pride and emotions (maybe).

High proportion of agricultural machines are already equipped with huge number of sensors to track down the performance of the machines but these are mostly used in the test environment but not exploited well enough by OEMs yet. The reason being lower connected machines, higher data transmission cost, huge data storage and processing cost and so on. As the necessity for predictive maintenance increases, the need for connected machines are increasing. The next step is to breakeven the additional cost of connected machine with the value generated by data and insights. Gradually move towards profitability by providing tailored warranty and maintenance package, identify early failure and diagnose, just-in-time cross sell and upsell of parts, reduced warranty cost.

Focus for near future

There is great awareness about the value generated by telemetry data from the machine. Great progress is made to collect the right data in most optimized manner and also to start understanding the data. These are stepladder to enter ‘Door of infinite possibilities’

Post data collection and storage are structured, below follows:

1. Reports: Report offer descriptive point of view on features and behavior with other related features. In addition, helps understanding the data outliers which pinpoints the error in data, sensor tuning, faulty parts, replacement time of maintenance parts.

2. Understand outliers for critical sensors: A report on value of sensors and its abnormal behavior will identify the area of focus for field team to focus on the possible upcoming failures. A dashboard of critical outliers will help monitoring machine anomalies.

3. Important components: A component level monitoring is the next level of intelligence extracted from telemetry data. For example when we are looking at the failure of catalytic converter there is a group of sensors which can indicate the failure like, inlet temperature, outlet temperature, NO2 conversion, Frequency of DeSOx, Temperature of trap. All these parameter give a hint at the deterioration of catalytic converter and will eventually lead to failure. Similar parameters for critical and high valued parts will educate to make an assessment of component deterioration.

Door to infinite possibilities:

A leap from here, to gauge all components together to arrive at machine health and reliability. This will deliver valuable insight with various application across organization. Parts replacement indications helps aftermarket to up-sell failing components and routine maintenance parts (personally this is one of my much loved use case). Machine health is the indication to reliability of key components and sustainability which eventually assist defining precise warranty pricing. Critical component failure open up the door for newer opportunities in the space of preventive maintenance and eventually predictive maintenance (decade long jazzy term!)

It doesn’t stops there, Machine usage helps sales team with opportunity identification and customer retention, increases customer-centricity by understanding the demands of customer, helps in recommendations for both new products and aftermarket. Also, helps building the stronger tie between aftermarket and R&D with quicker feedback loop.

Significant pullbacks:

Customers with willingness to signup will be relatively lesser till it becomes a new normal across industry, hence number of machines with active participation will remain lesser in the beginning. The entire journey will be a long-term investment for the organization and till break-even. Never ending learning which always comes back with surprises because of it’s down to the wire nature. Overwhelming information from telemetry should have a well-structured downstream system across organization to consume actionable insights coming out of it, else entire value chain will fade away eventually.

Closing note:

Personally it will be really interesting for me to see how agriculture industry move forward in next half a decade and how well integrates with technology, IoT, big data and cloud computing to improve current day agriculture. Awareness in customers will have an imperative role in adaptation of future solutions and organizations has a never ending journey in a fairly unknown ground.



Data scientist with an intent to make world a better place