Production has become a complex issue, even more than it used to be. Modern manufacturers deal with decreasing product life cycles, increased product complexity, rising global market interconnectedness and a high demand for customization, just to name a few of the challenges faced today in the manufacturing industry. That probably explains why, astonishingly, 70% of the companies listed on the Fortune 1000 in 2008 have perished since then.
In today’s complex economy, conventional manufacturing systems lack the flexibility necessary to respond to quick and continuous changes in market conditions. To survive, manufacturers need to develop new skills, invest in new technologies and embrace new business models. In other words, they must join the fourth industrial revolution, or Industry 4.0, by integrating the Industrial Internet of Things (IIoT), artificial intelligence (AI) and other digital technologies such as big data, analytics, virtual reality and robotics, in their factories.
This is clearly not an easy task. According to Accenture, only 26% of US senior executives believe their people are ready to work with intelligent technologies. However, Forbes also reports that 95% of them believe AI will have a critical influence on their responsibilities in the near-future. The first step manufacturers can take to reduce their gap in digital skills is understanding what production intelligence is and what benefits manufacturers can expect from it.
What is production data? What is production Intelligence?
Production intelligence, also known as enterprise manufacturing intelligence or manufacturing intelligence, can be defined as business intelligence for production management. It is an advanced framework aimed at analyzing contextualized production data to extract actionable knowledge and drive business results. Now, you’ll probably be wondering what “contextualized production data” actually means. Let’s proceed with order.
Production data refers to all the information that is persistently stored and used by professionals to conduct business processes. It is all the data that a business continually stores and manages to understand what is going on in its inventory, machinery and production system in terms of process, quality and efficiency. As such, in order to actually be meaningful and helpful to a company’s decision-makers, production data must be accurate, documented, constantly managed and, most importantly, contextualized.
Production intelligence and big data analytics in general require data to be contextualized because data without context doesn’t provide any trustable knowledge. Only once data is put into context, meaningful multivariate data relationships can be spotted and used to expand and refine the internal knowledge of the company. On the contrary, data without context might even lead to incorrect strategic decisions.
Let’s make an example. Imagine you are looking at production outputs and your data says “2”. This number tells a very different story whether you are selling space rocket engines or cars. The same data in one case suggests the company is thriving, in the other it means it’s going bankrupt. In different contexts, the same data can provide stakeholders with completely different insights. This example probably sounds a bit silly and reality clearly is more complex, but I hope you get the idea of why context is so critical in big data analytics.
Production data can be extracted from a wide variety of sources and can be structured in many different ways. For instance, production data includes operational data, such as the capacity, run times, and faults of production systems and related systems. It also includes ERP data on stock, orders, and due dates. Manufacturing intelligence allows to combine different kinds of production data to accurately visualize the entire production chain on a clear dashboard.
Once production data has been properly collected, contextualized, integrated and analyzed it can be used to help process engineers and plant managers monitoring key process KPIs, identifying new business opportunities and becoming data-driven decision-makers. For instance, production data can help understand and optimize inventory turnover, production waste and downtimes, production and inventory costs or working capital requirements, just to name a few examples.
The 5 features of production intelligence
Now that we defined what production data is, we can refine our understanding of what production/manufacturing intelligence actually means. First of all, we need to stress that production intelligence is not a software solution. It is a goal-oriented approach that requires software solutions to reach its goal of creating a smart, constantly optimized production system. Once this has been clarified, we can focus on the features that characterize this approach:
- 1. production intelligence is an integrative approach: it forces companies to integrate their ERP, MES and best-of-breed systems to look at their operations in a comprehensive manner that boosts overall utilization of those previously isolated software systems;
- 2. production intelligence considers production sequences as a whole: this feature is closely tied with the previous. The ratio behind this is that, by harmonising sub-processes and eliminating inefficiencies in the transitions among sub-processes, the uniformity and capability of the whole production process will increase;
3. production intelligence represents a framework for the interdisciplinary development and co-ordination of smart production processes: in simpler terms, production intelligence requires manufacturers to look at different fields of knowledge in order to optimize and standardise methods, technologies, and resources. This helps reduce the level of ambiguity through the production process while boosting performance quality. However, to successfully develop smart production processes, businesses must establish special procedures that ensure the involvement of all the necessary skills and resources at every stage of concept development;
- 4. production intelligence fosters continuous innovation: this feature is a consequence of the interdisciplinary and integrative nature of this framework, which naturally encourages people to learn from their colleagues with different areas of expertise. Such cross-pollination of ideas stimulates innovative ideas that can be used to improve technologies, methods and processes;
- 5. production intelligence promotes continuous learning within the entire production system: as we mentioned previously, production intelligence is a goal-oriented framework and therefore involves the identification of some key performance indicators to keep track of progress. Indeed, who would expect to understand if their strategy is working without accurately measuring results? Well, apparently, production companies, since many of them still base their decisions on their manual analysis and their managers’ intuitiveness, rather than on the insights from production data analytics. While, technically, this type of analysis could also be done manually, through Excel worksheets, manual data processing for the huge volume of data required for Industry 4.0 is realistically impossible.
- In contrast, manufacturers that embraced big data analytics and manufacturing intelligence have more control over the complexity of their operations, are able to perform need-based performance analyses and, eventually, can identify the optimal decision to make in that context. Overall, these three connected elements promote the development of a culture of continuous improvement that stimulates workers to constantly educate themselves.
PI benefits and conclusion
Production intelligence entails many advantages for manufacturers. It makes production processes more flexible, allowing factories to adjust their workflow based on market conditions and accurate forecasts. It reduces concept development and production cycle times. It improves quality, safety and costs control. It fosters a data-driven culture that encourages workers to continuously learn, innovate and improve.
As such, production intelligence can be no longer seen by manufacturers as a judgement call. While data security concerns and high upfront investment costs are some of the main barriers to the adoption of manufacturing intelligence, it has become a necessary choice for any production company willing to stay relevant in the next ten years.