One of Microsoft’s few failures over the years has been the Microsoft Zune, a MP3 player that crashed and burned – largely because by the time it was released, Apple had already cornered the market with its fruit-emblazoned counterpart. This is just one example of how critical time to market is.
Image Credit: Intellegens Limited
The crucial nature of time to market is not a new concept, but one that has guided global commerce for thousands of years. Thirty years ago, a classic McKinsey study showed that companies lost 33% of their after-tax profit due to delivering their products six months late.
Many factors affect time to market, such as distribution, marketing strategy, production processes, and supply chains. Driven by the success of Japanese manufacturing in the 1970s, the McKinsey study was part of a wave of interest in just-in-time approaches.
This article focuses on the first part of the value chain, R&D. A 2015 Boston Consulting Group report found that fast innovators achieve first-mover advantages, increase market share, reduce development costs, and improve forecast accuracy for their businesses.
R&D is therefore a race. In materials, chemicals, and manufacturing, it’s one of those where there’s a lot of opportunity for smart, innovative companies to push forward. Lucideon’s Dr. Richard Padbury cited another McKinsey report in a recent webinar, which found that there was an average development time of 20 years for new material.
How can machine learning (ML) change this picture? There are three main areas to focus on:
1. Perform fewer experiments
ML can help target experimental programs, which are still the biggest waste of time in many product development. ML helps select the set of experiments that most effectively explore potential solutions – or in other words, it can continually answer the question, “Which experiment should I do next?”.
Faced with such challenges, applied ML specialist Intellegens aims to reduce the experimental workload by 50-90%. To succeed, Intellegens had to overcome two major obstacles. First, he had to develop ML methods capable of extracting information from experimental data, even when that data is noisy and sparse. Most ML methods fail with such data. Second, it is also very important to accurately quantify the probability of success of a particular proposed experimental route – this allows for rational decision-making.
2. Increase the likelihood of breakthrough information
Breakthrough insights – the bringing together of critical pieces of information or moments of inspiration – are responsible for accelerating R&D. Although such breakthroughs can never be guaranteed, their probability can be increased. Here, ML can be invaluable, finding correlations that a human may never spot when exploring large, multi-dimensional datasets.
Recently, an interesting example emerged from a publication on antimalarial drug research. Using ML found a very promising candidate compound – but it was one the chemists admitted they would have dismissed as “misguided”. Subsequently, the compound was synthesized and its potency was experimentally validated.
Image credit: Intellegens Limited
Focusing on the word “overview” is also essential. Instead of running an algorithm and predicting an outcome, users need to understand Why the prediction is produced. Analytics like the importance chart (Photo) can help with this understanding.
3. Reuse of existing knowledge
“We would be so much faster if we only knew what we already know.”
The focus is on computer systems that capture legacy data and its interrelationships within industry. Examples within R&D may include results of experiments and analyses, results of computer simulations, and process data for chemical, material, and biological systems. Further downstream, there may be, for pharmaceutical products, data from clinical trials or patient data, or, for manufactured products, manufacturing or in-service data.
Again, ML can be applied to exploit these data resources. However, data scarcity remains a challenge: since data collected from multiple sources will inevitably have inconsistencies and gaps, it is important to have ML methods that can deal with them.
An ML model can become a valuable way to capture information and knowledge for later reuse. However, it can be difficult to apply these models in real business environments if running or modifying them requires significant coding or scripting knowledge or data science expertise.
Intellegens has found that many of its customers want to deploy the results of ML studies to their scientists and engineers via accessible platform technology, such as the Alchemite™ Analytics web UI.
Want to try something new?
If R&D can be characterized as a race, it is perhaps less surprising that it requires certain tactics and strategies to win – acting as an advantage over the competition. Machine learning could be that difference.
The previous difficulties in the successful application of machine learning are generally due to the factors discussed in this article: the need for quantification of uncertainty, sparse and noisy data, a lack of effective analyses, or perhaps the difficulty of deploying useful models. If any of these factors seem relevant, contact Intellegens today to find out how these challenges can be overcome before the competition begins to close.
This information has been obtained, reviewed and adapted from materials provided by Intellegens Limited.
For more information on this source, please visit Intellegens Limited.