For all intents, its manufacturer can no longer view a tractor used on a farm for a one-time sale. Across its ~22 year lifespan[i], the tractor will continue to need maintenance that could very well add up to more than its original price tag. Can manufacturers look at such lifetime value of customer relationships? The answer is “yes”. The Internet of Things (IoT), with its 50B-connected devices by 2020, is enabling it. Manufacturers of farm, construction, transportation and mining equipment, industrial products, white goods, etc., are already being assisted by the IoT to gain real-time views of the location and health of their assets.
Additionally, manufacturers can geotag assets such as tractors and combine asset usage data with anything else, say a GIS database. This combination can provide heat maps of where there is momentum in the usage of assets or cluster problems to help determine geo-specific solutions. In effect, the greater the number of data sources brought together, the richer will be the insights and the higher their value.
Some benefits are obvious: Predictive maintenance to boost uptime, reduce service costs, add to product lifespans, and increase customer satisfaction. Our research shows that an industrial equipment manufacturer with annual revenue of over US$10 billion could reduce the cost of service alone by US$140 million[ii]. This is just the tip of the iceberg. A larger opportunity awaits the entire ecosystem.
One of our clients, a major power provider, makes an ideal example of the exciting new opportunities. This client wanted real-time location data for every Backhoe used in a city. The data is used to identify Backhoes digging in the vicinity of a power cable or other underground equipment. A pro-active alert is sent to halt work before it caused damage to equipment. Providing the data built an additional revenue stream for the Backhoe manufacturer while the power provider prevented disruptions and repairs.
PARSING DATA FOR THE ECOSYSTEM
As the number of intersections between data sources grows, so does the opportunity. Imagine a product like a modern oil tanker or a container ship with a life span of decades. Such a vessel would be spewing out a torrent of data that can be parsed for use by multiple partners. The health of its pumps, turbines, crankshafts, etc. would be of interest to the OEMs and insurers; maritime agencies would be keen to know the exact location of the ship along with weather conditions: the condition of the cargo would be of interest to the seller and the buyer; while the overall health of the ship would be closely monitored by the financier. As is evident, a large ecosystem stands to benefit.
To make this happen, the volume, velocity, and variety of data being produced by ships, earthmovers, hospital equipment, nuclear plants, industrial appliances, etc., must first be tamed. Only then can it be refined into intelligence and services.
Traditionally, data has been captured in siloes. Some familiar siloes include ERP, CRM, and even spreadsheets. To compound the problem, each region of a business could have its own instances of these siloes. Over time, especially with IoT data, the points of integration multiply. Managing the data becomes an impossible task, and exploiting it becomes impractical.
One solution is to have a data lake where data is held in its native format until it is required. But this too poses several obstacles. For example, if the frequency of refresh goes up, it can choke the network. This calls for an innovative solution—a consolidated real-time record of all connected and non-connected assets with a metadata layer that indicates where the required data is located.
A PLATFORM TO INTEGRATE DATA
The system would comprise of four layers (see Asset Management Platform):
- Edge Device (ED): Edge device picks up data. The device would have its own information management capability that would permit it to determine when to transmit data based on urgency, need and network availability.
- Data Shards (DS): With IoT, things get tricky very quickly. Machines generating the data could be running in China or Germany and be governed by local regulations. Thus, discreet shards would be required for locations, business units, product lines, etc. This layer would also capture data from enterprise systems.
- Hyper Dimensional Modeling Framework (HDMF): This layer manages the relationships between assets, owners, users, organizations, partners, etc., and determines the type of data each asset can create, measure and communicate. It also determines how data sources can be mashed. It is essential that this is executable in a scalable manner across millions of assets.
- UI Application Framework (AF): This layer enables the API gateway which exposes aggregated and computed data to users and applications.
To use the earlier example of a shipping vessel, this type of platform would ensure that owners and operators would be able to monitor and share data across their fleet; OEMs and service providers would be able to access data that would help them predict component failure[iii] and improve the design and performance of their products; and create contracts between actors in the ecosystem for data consumption[iv].
OPENING THE FLOODGATES OF OPPORTUNITY
There are several interesting possibilities, aside from monetizing the installed base, that an asset management platform would give rise to. Equipment like excavators, compactors, drilling machines etc. could be charged based on the number of hours they are operated and the health of the machine rather than for the number of days they are rented[v]. Wipro had an interesting request from one customer of our IoT-based asset management platform. The customer wanted help in tracking and recovering stolen equipment (Yes, we were successful. The machines were tracked to a location several hundred miles from the locations where they were deployed). Equipment manufacturers will be able to sense the need for consumables from owners, create stocks and make optimized and customized offers[vi]. And finally, manufacturers will be able to identify equipment that needs intervention and therefore predict the quantum of work and reduce carry costs.
For a technology company like ours, implementing asset management solutions and connecting devices, generating data, standardizing formats, etc., has been the easy part. Technology can achieve this. The real problem has been user adoption. The technology quickly becomes useless if the user cannot find appropriate value in it. This is why before opting for asset management platforms it is critical for an enterprise to examine what the system is going to deliver and how important it is to business.
[iii] They can buy data from fleet owners
[iv] Wipro’s cloud-based asset management platform Looking Glass is designed to provide these capabilities to organizations across industries. The roadmap for Looking Glass includes Data Discovery Platform, Asset Performance Analyzer App, Partner Value Enhancer App, our AI solution Wipro HOLMESTM and a Digital Ecosystem Driver App
[v] Real-time usage data would guard against revenue leaks
[vi] This would mean reducing campaign costs which otherwise tend to inflate because of board-based and imprecise marketing approaches