The term “quantamental” is a mash of quantitative and fundamental, conveying a mix of computing power and human expertise. The expression originated in finance, to represent the convergence of quantitative analysts with investors focused on fundamentals. It’s also the new model for digital manufacturing—an amalgam of quantitative analytics powered by AI and machine learning and the classic, fundamental soundness of meticulous process planning and lean manufacturing strategy.
Quantamental manufacturing has brought a new toolkit to classic production—just as quantamental investing has done for finance. Quantitative investing relies on mathematical and statistical modeling of huge amounts of data; fundamental investing is what you see from people who pour over operational details and balance sheets with a fine tooth comb.
Quantamental manufacturing operates the same, turning loose quantitative computer models, AI and machine learning onto big data sets to uncover patterns 
and make predictions. On the fundamental side, seasoned shop-floor practitioners pick apart the ins and outs of the production process with a combination of hard-won common sense and the precepts of lean manufacturing.
Lean manufacturing emerged in the late 1980s from the drive to wring excess from the production process; it was based on the legendary Toyota Production System (TPS), with its relentless pursuit to eliminate waste. The first “Kaizen” (continuous improvement) teams spent prodigious amounts of time examining processes, gathering data and recording their results on clipboards. Enterprise Manufacturing Intelligence (EMI) software then let companies extract this data automatically from plant control systems and databases; Manufacturing Execution Systems (MES) translated and distributed the data to the machine level, so operators could get a better grip on processes and follow lean instructions.
Quantamental manufacturing expands that computer-assist to encompass multiple physical assets, processes, plants, and ecosystem suppliers. Its AI and machine learning don’t rely on pre-programmed logic, which couldn’t possibly cover every situation and eventuality. Rather, they undertake self-learning analyses of the data to unravel how things work—and how things can work better.
Quantamental manufacturing also pulls in data from beyond the enterprise orbit. A quantamental manufacturer launching a new product, when projecting sales to assess production and supply chain needs, will not only factor in a machine learning analysis of previous clusters of products with attributes similar to the new launch item—it will also vacuum up social media sentiment and geospatial mapping data to better match anticipated demand with item build and inventory staging along the supply chain.
Quantamental manufacturing has upended the talent mix for maker companies. The ranks of old-school production salts on the plant floor are coalescing with algorithmic-driven colleagues with the skills to install a quantitative methodology that helps make sense of the virtually immeasurable flood of data now flowing through, and pouring out of, the production process.
Quantamental manufacturers balance these human and machine-based approaches to react as quickly as humanly and “machinely” possible to today’s whirlwind of unceasing product innovation and shifting market preferences, where it’s imperative to identify issues right away and make changes on the fly.
John Martin writes about technology, business, science, and general-interest topics. A former U.S. correspondent for The Economist (Science & Technology), he writes for the private sector, universities, and media, and can be reached at jm@jmagency.com.