The implications of big data in manufacturing: How to super charge real-time connected supply chains

Logistics is the latest industry to face a bold makeover in the Singapore government's $4.5 billion Industry Transformation Programme. The initiative aims to achieve a value-add of S$8.3billion (US$6 billion) and 2000 jobs for professionals, managers, executives and technicians (PMET) by 2020 through leveraging on new technology such as big data and analytics.

The explosion of supply chain-oriented big data is tremendous, and poses great potential for the professionals to create a real-time connected supply chain. From a global perspective, analysts predict 30-times growth in connected devices by 2020. 26.9% growth for IoT in manufacturing thru 2020 (Forbes), 13.5% compound annual growth in connected trucks through 2022 (Frost and Sullivan) and growth in RFID tags from $12 million to $209 billion by 2021 (McKinsey).

Locally, a study by audit firm KPMG for the Competition Commission of Singapore found that logistic services in the country are increasingly starting to adopt more data analytics to improve business services. The study took note of how data is used to monitor drivers’ driving patterns, predict customer demand, and optimise routes, among others. Parsing data was found to be beneficial, particularly in forecasting customer demand, reducing delivery cost, and reducing errors.

The impact of these trends will be huge: organisations will soon be awash in all the real-time big data necessary (coming from devices, sensors, vehicles and long histories of operational transactions) to transform their supply chains. The question is not whether this will happen - it’s what impact this trend will have on your company and, more pointedly, what can your organisation do about it?

Impact across the supply chain
We have already witnessed several examples of real-time, connected supply chain processes being implemented across all pillars of the extended supply chain: design, purchasing, manufacturing, distribution and marketing and sales.

In design. Leading companies are increasingly leveraging vast volumes of social big data to understand product requirements, “math-based” big data to drive virtual and 3D printed prototypes and sensor big data to drive digital test simulations.

In purchasing. Enterprises are analysing long histories of sourcing event data to identify exactly those variables (i.e. time of day or year, number of suppliers invited, energy costs) that led to the lowest cost sourcing outcomes and then quickly applying this knowledge into current day sourcing practices.

In manufacturing. Practitioners are collecting and analysing shop floor sensor big data to monitor real-time operational performance, discover optimal process parameters to maximise quality and yields and predict optimal maintenance intervals for equipment.

In distribution. Professionals are increasingly analysing logistics big data (i.e. GPS, RFID, traffic, weather) to dynamically re-route trucks and optimise the design of their distribution networks.

In marketing and sales. Real-time analysis of demand big data (i.e. social, web logs, POS, consumer location) is providing the ability to understand and predict consumer needs and actual demand, while analysis of long histories of marketing campaign data is providing the ability to identify the key marketing variables driving effective marketing outcomes.

How you can supercharge your supply chain
With big data impacting so many supply chain processes, here are some steps to start.

Begin with aligning big data to your company objectives, by considering your goals and targets. For example, if increasing revenues is a high priority, consider a design- or marketing and sales-related big data use case. Conversely, if cost reduction is a major focus, consider use cases across the purchasing, manufacturing or distribution domains.

Once you have outlined your objectives, identify your “line of business” champions. We have found that technology architects, however enthusiastic they may be regarding big data technologies, often have difficulty getting supply chain initiatives approved within their companies. The advice we offer them is to identify and join forces with business process owners who can serve as champions for big data transformation projects moving forward. Without such line of business support, selling big data transformations is an uphill battle.

Next, identify the state of your data. How accessible is it? A great initial step involves creating a “data lake” that will support future supply chain transformation initiatives. Without your ecosystem data under management, it’s difficult to move on.

Starting with small projects will make the journey more tangible, digestible and realistic. Resist the impulse to jump to the most complex use cases immediately. Choose small scoped projects that will give greater visibility into your supply chain. The lessons learnt from small projects will make the huge, transformational supply chain big data programs much easier and faster to achieve.

Supply chain use cases generally range from visibility-related (simpler) to optimization-related (more complex) examples. Often, just gaining basic visibility to supply chain information (i.e. process monitoring or inventory location tracking) can provide immense value, without the need to resort to more complex optimization use cases (i.e. quality/yield optimization or predictive maintenance).

The race is on
In short, we see the influx of real-time big data and high-performance analytics enabling new levels of supply chain performance, underpinned by supply chain visibility, performance monitoring and the ability to optimize current and future actions based on lessons learned from the past. Industry leaders are already building, evolving and benefiting from their big data foundations, so don’t wait; the race is on.