-

AI Operations Optimization

Correlations

Problem:

What are the drivers in a business?  Most don’t quantitatively know.  For example, a manufacturer thinks providing its distributor with a cash incentive for each sale of the manufacturer’s product will drive more sales.  Did it?  The manufacturer did not know but did such because its competitors were doing such.

Many actions in promotions, pricing, product mix, inventory, and so on are made on beliefs or competitive responses without knowing quantitively if there is a relation between the action and the desired result.

Cost:

Excessive promotions, lowering of prices, carrying of products lead to an inefficient use of capital.

Aurora Solution:

Built-in correlations enable point-and-click application on any field(s) of data at any dimension(s) to automatically quantitatively identify where relations exist and don’t exist. Instantaneously apply internal and external correlations to confirm or disprove a “belief”.  In the example above, the manufacturer was “half” right; i.e. in the first half of the year the incentive had no relation with sales, but in the second half there was a strong correlation.  Thus, the manufacturer should significantly lower first half incentives and increase second half, and the capital reduction in the first half could be shifted to other incentives that do have strong correlations to sales in the first half.

Correlations also find disconnects in supply-chains between materials to manufacturing to distribution to retailer.

Applications include supply chain, product portfolio management, promotions, pricing…

Optimizing Costs

Problem:

The “middle” of the P&L is riddled with costs that get only the most basic of scrutiny; e.g. commodities in manufacturing are set as a fixed cost; guess too high and manufacturing is stealing money from marketing, but guess too low and production margins are reduced. Another example is freight, how do you know if it’s too high or too low; just a price or SLA (that often isn’t measured) doesn’t reveal much.

A universal example of excess cost is headcount.  All managers cry for more but there is little quantitative push-back, because, well, the mangers are in the thick of it and know best.

Cost:

Setting commodity price is typically wrong because it’s fixed, which means it’s like a broken watch – right only twice a day. Focusing on the price of freight without knowing the “efficiency” of the price often misses the real low-price carrier.

These misses, and others, typically translate into an average of 5% added costs, which eats at the margin.

Aurora Solution:

Commodity pricing can be tackled with AI-Enabled forecasts that often are better than industry average pricing.  Further, AI-Enabled forecasts are dynamic, so manufacturing doesn’t have to use a fixed price, thus freeing capital for marketing.

For other costs, the KPI (mean) value is built-in.  When used in conjunction with the built-in Efficiency Driver, apples-to-apples comparisons can be made that yield a quantitative comparison of costs vs. a misleading comparison based on the absolute value of costs.

Efficiency Opportunities with Statistical Drivers

Problem:

Business knows its revenue and expenses, and most comparisons are made on these numbers. Important, but only half the story, as these comparisons often are not on an apples-to-apples basis. For example, Department A spends $1m and Department B spends $2m, then if both departments are similar, A is better than B based on costs. But, not necessarily so, if an “Efficiency Driver” of, say, FTE is introduced; i.e. a denominator to the cost to make a normalized comparison. Now, if A is $1000 per FTE and B is $750 per FTE, then B is better than A, because it is more efficient with its costs.

The same analysis applies to the top-line. If Product A has more sales than Product B, it is not necessarily more efficient in generating sales; i.e. if A has sales of $750 per unit sold and B is $1000 per unit sold, then B is more efficient to generate revenue than A.

Not knowing the comparative efficiency leads to missed opportunities to reduce cost and increase revenue.

Cost:

The typical business can improve profit by 5% to 10% by better knowing the efficiency of its revenue and expenses.

Aurora Solution:

Efficiency drivers can be selected and applied with a point and click to enable apples-to-apples comparisons on any field(s) of data at any dimension(s).  Further, the built-in financial analysis of the impact of the trend of the efficiency on costs and revenue automatically finds, quantifies and prioritizes areas for expense reduction and revenue gain.

Applications include product portfolio management, headcount planning, promotions...