Why industry 4.0 technology works best in bundles

Where the factory of yesteryear brings to mind steaming furnaces, greasy machinery and the clang of chains, today's manufacturing is about embedded sensors, digital twins, autonomous vehicles and robots.

Indeed, entire sectors have undergone a transformation of processes, products and business models, which are often affixed with the term “smart” (the ubiquitous new catchword of this paradigm shift).

At the core of this Fourth Industrial Revolution (4IR) transformation are digital technologies such as the Internet of Things (IoT), cloud computing, big data and analytics, as well as additive manufacturing.

There is substantial agreement among scholars about the positive impact of 4IR adoption on corporate performance. For instance, benefits can be measured as reduced operational costs, higher capacity utilization rates and improved product development times.

But smart business solutions should involve more than just adopting disconnected technologies, as we point out in a recent research paper co-written with Francesco Galati, Margherita Molinaro and Elena Pessot. Instead, we suggest “bundling” technologies together.

This helps achieve synergies, leading to major business improvements. Yet despite the vast emerging literature on 4IR, the complementarities between different 4IR technologies are not well understood.

This is why we set out to understand how combinations of 4IR technologies – and their functionalities – can be realized in multiple ways, aiming to better unleash performance outcomes beyond single technology adoption.

Systematizing the 4IR technologies considered in the literature, we thus identify them as visualization technologies (e.g. smart glasses), computing technologies (e.g. modelling solutions), network and sharing technologies, digital production process technologies, and data-processing technologies (AI, big data analytics).

While previous work has studied “bundles” of technologies, identified based on their application areas (smart manufacturing, products, working and supply chain) or production targets (such as flexibility or process quality), a detailed overview of the potential patterns of such 4IR complementarity was still lacking. So, adopting the holistic perspective of systems theory, we did just that.

To examine the complementarity of 4IR technologies and their performance impact, we studied the large and vibrant 4IR ecosystem being developed in Italy via interviews with 13 experts in the field and more than 150 use cases. Overall, our results show that 4IR technologies can impact a firm’s performance when adopted both as stand-alone technologies and as bundles, but major gains are more evident in the latter case.

The adoption of most classes of technologies, even if not combined with others, may still positively influence performance. Only computing technologies never emerged as an effective stand-alone adoption.

For example, the experts surveyed cited several times the IoT to create smart products, which can enhance customer service and improve the perceived product quality. In the data-processing group, two technologies were discussed as stand-alone effective, namely big data and analytics, and artificial intelligence.

The former can increase the overall equipment efficiency in the manufacturing process, and, as a smart working application, reduce demand planning efforts, whereas the latter can improve production processes, for example quality control, with AI algorithms analyzing photographic images of the products to quickly identify possible defects.

The second result shows that 4IR technologies may be combined in 10 different bundles to increase performance. Such bundles are underpinned by two main groups of technologies, which seem to play a pivotal role in all the other combinations: network and sharing technologies, and data-processing technologies – fundamental building blocks operating as platforms for the other 4IR technologies.

Combined together or with the other technologies, they affect 33 performance dimensions; the most frequently cited outcomes are error reduction, product quality increase and increased worker safety.

We detail one case of the adoption of network and sharing and data-processing technologies: This bundle can support the management and control of a production process, thanks to the collection, storage and advanced analysis of manufacturing data, thus improving almost all the performance categories, including energy consumption.

Competence Centre A (where one of the experts of the study serves as manager) recently supported a firm operating in the automotive sector to diagnose energy inefficiencies in its industrial process by leveraging IoT, big data, and analytics. The solution has been installed into four automative plants, leading to a 5% energy saving (quantified as €40 million).

Drawing from our results and considerations about “platform” technologies, we identified three kinds of complementarities:

We draw two conclusions. From a theoretical viewpoint, the complementary approach exceeds the sum of the impacts reached with the single technologies, enabling benefits not reachable by scattered adoption. From a practical viewpoint, the framework informs managers which technologies provide a specific outcome, for them to avoid technological adoption based on guesses, gut feeling, or experience; often, this results in a weak decision-making approach. In particular, platform technology bundles represent the solution with the highest versatility.