Operations Analytics: Why it should be in your bottom-line strategy

CGN was established on the predicate that operations are at the heart of any company. Operations are responsible for ensuring revenues, delivering customer satisfaction and, above all, managing the largest chunk of cost. Managing operations to drive business performance has been a focus for decades. Some years ago, it seemed that most significant improvements had been achieved and that only incremental improvements were possible. Today that has changed.

A combination of information technology advances in pervasive IT deployment, networking and the Internet of Things has made data widely and easily available. The drop in cost of number crunching by servers and access over the Cloud has made analytics a keystone for the next level of operational improvement.

A couple of examples will make it easier to understand.

Distribution of consumer products

Most FMCG companies have moved to a replenishment model of differing levels of sophistication. However, all of them face an operational challenge.

Replenishment models are deceptively simple. If the stock level of the next link in the chain drops below a norm, a replenishment order is triggered. The real challenge lies in setting the “norm”. A national FMCG company with 400 SKUs servicing 20 warehouses will have 8,000 norms to be set at a primary level. If there are 500 distributors, there will be 200,000 norms to be set. How will this be done for a new product with no sales history? Or, adjust stocks for a four-week promotion? This problem arises wherever past history is not a valid basis for estimating demand at such a granular level. How frequently can planners sit and rework these norms? Increasingly today, the proportion of sale for which historical data is not available is increasing. Historical data, in short, has become history.

This problem is worked around by becoming more responsive. Just as a tennis player who has brilliant response time can overcome a new and unknown opponent, so can a responsive company adapt quickly to marketplace changes and behaviour. Operational analytics allows high speed response based on quick collection of granular data by applying appropriate algorithms tuned to industry characteristics.

In a subsequent article, a case study of moving from a monthly to a weekly planning cycle based on data driven planning, we will see how shortening of planning horizons gives improved benefits. If those planning horizons are dropped to a daily or twice daily frequency, the next level of improvement in fill rates and inventory optimisation can be reached.

Shop floor cost reduction

Customers want deliveries quickly and value responsiveness, but they also want them cheap and really like lowered cost. Cost reduction is a continuous exercise on every shop floor and managers are always looking for ways to bring cost down. Operational analytics opens up new opportunities for improvement in cost management.

At its core, operational analytics allows for two methods of identifying opportunities for cost reduction. Firstly, it can take huge volumes of machine data that would simply overwhelm manual analysis and, secondly, it can apply correlation techniques to identify possible causes that can lead to new improvements.

Take the case of energy consumption in a process that is churning out products by the thousand in a shift. By studying energy consumption on a machines through each cycle, for every cycle, it is possible to identify a wrong setting that is consuming excess energy, or time being wasted in a cycle when energy is still consumed or even a particular material that is leading to excess energy consumption. Where shops would average data over weeks, shifting to averaging data over shifts threw up opportunities for improvement. Then, in experiments, studies would be made where data was averaged over each cycle to identify anomalies. Now, it is possible to study data over minutes and seconds, throwing up opportunities to reach new levels of precision and reduced waste.

At CGN we are looking into how data analytics can be applied to supply chain problems across different industries. Most work on this area has been on looking at micro markets and identifying sales opportunities. However, we believe that operational analytics can have a significant impact on both responsiveness and efficiencies, if correctly applied.