Manufacturing-As-A-Service: A New Business Model


It’s time to redefine manufacturing as a service.

As more manufacturers evolve their business models to take advantage of IoT technologies, the concept of MaaS is itself evolving to include not just mass customization but also a growing list of product-based services.

To some, MaaS means the ability to manufacture one-off or highly customized products. In consumer electronics, that can apply to storage capacity, processor speed or screen size, or, as Protolabs’ CEO Vicki Holt notes, even more innovative features, “…such as exact-fit headphones and ear monitors, custom-colored keyboards and joysticks, even electric razors customized to the fit and shape of your hand.”

To others, MaaS means making products that are able to deliver ongoing value to customer and manufacturer via built-in sensors and Internet of Things communications. For instance, Caterpillar now outfits its machinery to report through the IoT on a range of equipment and operation functions that can help Caterpillar and its customers analyze field performance and optimize project usage.

Narrowing Customer Focus

As different as they are, these and thousands of other examples have three major facets in common. The first: the more the manufacturer can get closer to the customer, the greater the opportunity for success – and profit.

With mass customization, the better the manufacturer’s ability to build a personalized product, the more likely to please the customer and build loyalty. It’s much the same for a product-as-service scenario. The sensor-equipped product can now keep up a conversation with the manufacturer, providing up to the minute information on the specific product that is now in the customer’s hands.

The product-as-service conversation can center on performance, with IoT sensors measuring and reporting everything from environmental conditions to the performance and reliability of the unit’s parts and subsystems. And it can do much more. A smart TV can deliver customer viewing habits back to the maker, to sell to advertisers. A smart vehicle can deliver location information to a smart-city application that wants to predict traffic patterns around, say, a football game.

But in each case, as with mass customization, the value of the product is measured largely on the maker’s ability to identify and act on specific information about that particular product, and that particular customer.

An Audience Of One, But A Band Of Many

The second thing they have in common: the tighter the focus on the individual customer, the more complex the supporting value delivery system.

To make personalization work, the manufacturer’s entire operations ecosystem must play well together. Product design, manufacture and assembly have to synchronize with inventory and suppliers, with order processing and logistics, and with marketing, customer relations and field service.

Operations systems – manufacturing execution systems, ERP systems and their associated software applications – must be able to pull and process data as needed from a variety of divisional data stores. And, of course, speed and accuracy of operational processes are critically important. The customer doesn’t care about these machinations, but the customer does care about fast delivery and a responsive product, and about effective customer service and rapid repair turnaround. None of these are possible if operations performance is insufficient.

Needed: A Real-Time Analytics Overstructure

A need for an overarching fabric of real-time IoT analytics represents the third facet that MaaS manufacturers share in common. A well-run, up-to-date IT infrastructure is essential to success for most businesses, but it’s not enough to carry the manufacturer forward into the new world of MaaS.

he reason: IT likely will not be able to keep up with the conflicting pressures of boosting process speed and increasing systems complexity. By contrast, this is where an end-to-end real-time analytics fabric does its best work.

The analytics overstructure literally sits over the IT applications, drawing data from them and then comparing that data with other information coming from contextual data stores. It collects even data in real time, then uses machine learning models to discern various patterns that emerge. This way, it can spot trends – whether positive or negative – in time for the business to take meaningful action.

It can see, for instance, how a malfunction in one system may affect a different system located miles away. Or it can see how a delay in a materials shipment might affect a key customer. Or it can tell how best to re-route product deliveries around a sudden fire.

The analytics overstructure also collects real-time information from the sensor-equipped products in the field, alerting field service of the need for proactive maintenance, or customer service and marketing on a new cross-sell or up-sell opportunity.

This isn’t to say that a real-time analytics overstructure can replace the IT function. It can’t. In fact, IT can add value by maintaining data consistency and process governance for the teams – the data scientists, software engineers and business analysts – that are using the various analytics tools. In return, the IT group itself can get end-to-end process visibility, a feature of the real-time analytics overstructure, and can use this to troubleshoot potential problems or help with new application deployments.

This way, IT and real-time analytics should work well together to give the manufacturer the opportunity to define its own best approach to MaaS.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>