HomeConnectivityState of things | Smart manufacturing (part 4): Buying digital change

State of things | Smart manufacturing (part 4): Buying digital change

This article is the fourth instalment in a series taken from a longer report, Smart manufacturing: asset management, predictive maintenance, dynamic scheduling and other use cases, from August 2018.

The report tells the story of smart manufacturing in four chapters. The first two examine the expectations surrounding the IIoT movement (‘The hype and the glory’), and its application in the manufacturing sector to date (’An alternative truth’).

The second two chapters consider practical advice for manufacturers starting down the road to digital transformation, by turns offering guidance about how to approach the technology (‘Use case modelling’), and, how to fund and recover investments in digital change projects (‘Buying digital change’, see below).

The full report features additional information and use cases – as well as the article in its entirety. Click here to download the full report. See links at bottom for all instalments.

The fundamentals of operations management have hardly changed; the difference is the the digital tools used to make businesses slicker have gone up in their sophistication and down in their expense. The effect has been to “turbo charge” operations.

But what is the secret, where smart manufacturing has achieved a return? How is digital transformation bought and sold in the manufacturing sector?
Platform vendors have various commercial models. In his own round-up of smart city platforms, Owen at ABI Research placed ABB’s Ability proposition third, behind equivalent platforms from PTC and GE Digital, largely on the basis of its sales offer. ABB offers it free of charge, even despite its well regarded innovation, charging for applications and solutions instead.

ADLINK runs ‘digital experiments as-a-service’, commercialised as a subscription-based offer as ‘DXS’, to provide a “safe space” for customers to test out their digital hypotheses. “They get to discover where IoT will result in true digital transformation and high returns,” says Speed.

The DXS package combines edge hardware, data connectivity software, and services for a monthly fee. “We work with other vendors to create a solution that can be evaluated at a smaller scale and then moved beyond a proof into full production,” he says.

Hitachi promises a 3:1 return. For the Japanse firm, it is always the same conversation: it starts as an exploratory pursuit, in close collaboration with the client, before it finds it mark, and shifts hidden levers in their systems and processes. “It is all done in ‘co-creation’,” says Kinsey, adopting management-speak for collective problem solving.

“We run failure and affects analysis. We draw a heat-map of all the data. We seek to understand where the key data sources are, and how they crossover. We understand the process, and ask people why they do things. Eventually, we build a hypothesis,” he says.

Ultimately, the Hitachi sets a 3:1 threshold for its digital transformation projects; Kinsey’s team will not start work, proper, until its hypothesis is projected to achieve such a return. If the ROI does not stack up, then they should down tools, and start over. If it does, they should expect to “turbo-charge” their business, even if there is an initial lag as the analytics are hammered out.

“That’s the magic number; if a client invests €1 million, they should get €3 million out the other end, in 12 months. It’s a hypothesis; but we won’t start without it,” says Kinsey. A minimum viable data set is born of the hypothesis, and a minimum viable algorithm is born of the data set, presenting a logic to predict downtime and, therefore, to adjust processes and schedule the maintenance of equipment.

“It gets complex quite quickly,” he says. Hitachi tends to run two-week ‘sprints’, where the theory is developed and tested, and repeated again, until a fine-tuned algorithmic logic is achieved. Has the rate at which Hitachi achieves the 3:1 return accelerated, or has the ratio stretched to 3:1, with the increasing sophistication of digital tools? “The acceleration is just like with turbo-charged cars from 20 years ago; you hit the accelerator and there’s a lag, and then, boom, it takes off. It isn’t linear in digital either,” says Kinsey.

“You don’t necessarily get to the value as quickly as you did in the old days, but once you do, it goes way beyond. You have to go through this front-end process of experimentation – where you make mistakes, hit dead ends, pivot, and test your assumptions. That’s part of innovation. But once you have it cracked, you very quickly bring your plant to new levels of innovation and capability. It really is a turbo-charge effect.”

Beyond the initial boardroom discussion, to establish the scope of the transformation project, Kinsey’s team seeks first to engage the ‘wizards’ within each manufacturing operation. “If they have Black-Belts, we’ll engage them – because they understand the processes and the numbers, and have been running stats on the shop floor for some time,” he explains, in reference to the rank masters of the Six Sigma doctrine, which proposes a set of data-oriented management techniques to raise quality in manufacturing.

“Most European companies have a crew of them that work in operations. They are likely to be engaged in analysing process capabilities and process controls already. When upgrading to a digital operation, it’s better to build on the work that has already started.”

PTC says the same. “A key element, if not the first element, is to make sure manufacturers can leverage everything they already have in their factories,” remarks Provencher.

Gallant underlines this, with a sniff of fighting talk. “Not all solutions play as well with others” he says, making particular reference to its Kepware division, acquired in 2016, for industrial connectivity.

Gartner places PTC top, from Hitachi Vantara, in its review of smart manufacturing platforms (see full report); in a separate analysis, ABI also rates PTC’s portfolio as the most expansive, and makes notable mention of Kepware. “Everyone’s high on PTC, and for good reason,” says Owen.

The offer of compatibility with legacy systems, minimising capital expenditure, is crucial to the sale of digital transformation. Provencher gestures to another slide (see full report), a borrowed visual from LNS Research, which shows MESA International’s ISA95 model for manufacturing execution systems (MES), covering also enterprise resource planning (ERP), supervisory control and data acquisition (SCADA) and process control system (PCS) systems.

“On the left, you have all the traditional systems every factory will have, whether it is producing soap or airplanes. We focus on getting more value from them,” explains Provencher.

“You don’t have to move the data, or create this ugly data lake. You keep the data in that mass of systems, and, in a very flexible and modern way, you have the ability to retrieve it as you need it, and mash it up in new applications to gain new insights.”

This way of selling, of introducing an expansive and expandable starter system, is novel for manufacturing clients. The deployment of first use cases is rapid, he says. “Instead of having a two or three year replacement cycle, where you are trying to standardise the same ERP across all your factories, you layer ThingWorx on top of what you already have. You pick a use case and go live in five or six weeks, without disrupting or replacing anything.”

Gallant references Japanese car parts manufacturer HIROTEC, which sought to leverage predictive analytics to eliminate unplanned downtime entirely (see page 24). “HIROTEC put 62 CNC machines online, and found enough efficiencies within six months to avoid the purchase of two new multi-million dollar CNC machines. The work funded that, and then funded the next project, and the next,” he says.

Herrera at OSIsoft says the business case is clearer. “We often see companies achieve payback in two years or less,” he says. He has top-line examples of payback on its PI System: Australian utility AGL has recovered AUS $18.7 million in the three years; US gas pipeline company DCP Midstream recovered $25 million in one; Hungarian oil refiner MOL has added $1 billion EBITDA to its bottom line since 2010.

Theresa Bui directoer of IoT strategy at Cisco, switches the conversation to the running costs associated with smart manufacturing. Companies get the benefits of enabling more efficient operations and productivity, but get stuck on the new-style maths, she says. “One of the main barriers to IoT adoption is a clearly defined ROI. A clear understanding of the costs around delivering industrial IoT services is needed.”

Attaching devices to the internet, typically via sensors, radio-frequency identification (RFID), Wi-Fi, or cellular connectivity, and guaranteeing the labour to support them on the shop floor, in the field, and on the support desk is hard to grasp. The biggest drain is in operations, service reliability, security, cost management and scalability, she says, where a lack of automation and insight hobbles flexibility.

The pitch from vendors is that connectivity management platforms and data control solutions create order amid the new digital chaos. They afford service reliability through real-time diagnostics and robust controls for authentication and access, and remove the attritional hassle of launching devices manually, with zero-touch, error-free provisioning.

For his part, Kinsey is happy to argue the toss, if it is to be challenged. Hitachi’s 3:1 ratio has not been easily deduced, he says; it has been developed and tested over a decade, through the company’s own manufacturing, as well as its clients’.

“We’ll go up against our competitors, which either don’t have a business case, because they’re just selling technology, or will debate the average return is more or less than 3:1,” he says.

The point is the 3:1 model works to scope out the project and focus minds on both the potential and the viability of the proposed transformation. “It’s a starting point, a rule of thumb; in some cases, we go way beyond.”

We finish where we started, with Yost and Kinsey, in separate dialogue, spliced together like they’re two industry veterans chatting in a bar. Kinsey does not much care for the term ‘IoT’, he says. “It’s not a product; it’s an architecture. And I wouldn’t call it ‘IoT’, either – I would call it ‘digital’.”

Indeed, there is a sub-text, here, in this narrative around digital transformation, linked to the marketing of new technologies and applications. The terminology is confused, and the approach is wrong-headed. “There is a lack of a uniform definition of what IoT is, as it applies to a manufacturing domain, and a lot of uncertainty and confusion as a consequence,” says Yost.

It is the same with AI, a major trend in manufacturing, and a classic example of technological semantics gone awry – where one person’s definition, of sentient machines, is quite different from another’s, of advanced analytics. “A year or 18 months ago, no one was talking about AI in manufacturing. What is AI, anyway? Its truest definition is not what people are asking for at all. But manufacturers know there is something in it; they know there’s promise.”

Kinsey talks about work with a European car maker, and the hard grind of innovation. “Its culture was the very opposite of innovative – it was about compliance, and not taking risks. It’s the same for all these big automotive manufacturers in Europe.”

Innovation is not available to pre-order and take-away, he says; it has to be worked at, in partnership. Instead, the Japanese firm has looked to set up like a Michelin-starred industrial consultancy to transform manufacturing. “It is a learning curve for the client,” he says.

Everyone is wiser in the end; the systems and culture are better tuned for digital change. Yost rejoins: “These technologies have turned a lot of heads, but the market doesn’t get it from an organisational and cultural perspective. It’s a ‘people, processes, technology’ discussion, as always – where the tech is the driver. But as we mature, we will find that is backwards – that we still need to be driven by processes and talent. The technology has to fall in line.”

A full version of the article, including additional information and use cases, is available for download. Click here to download the full report. A new report and webinar on edge computing in industrial IoT setups, called AI and IoT at the cutting edge – when to move intelligence out of the cloud and closer to the action, is also available; go here for the webinar; go here for the report.

State of things | Smart manufacturing (part 1): The hype and the glory

State of things | Smart manufacturing (part 2): An alternative truth

State of things | Smart manufacturing (part 3): Use case modelling

State of things | Smart manufacturing (part 4): Buying digital change

Previous post
Editorial Report: AI and IoT at the cutting edge – when to move industrial intelligence closer to the action
5G manufacturing
Next post
HERE establishes AI research center in Austria