In Search of the Dashboard You Deserve

Most software packages show current views of current forecasts, sales, production, and inventory. But what if you wanted visibility of the underlying trends, changes, patterns in all systems:

Volume

High volume products with high revenue
Old, phase-out products before they become a write off (cost)
New products (phase-in) and when they’re going to hit (ramp up)
Volume available to strategic customers (by customer, by region, by warehouse)
Is manufacturing stuck making slow-moving products?

Velocity

Volume per unit time by warehouse, region, strategic customer
Are we manufacturing to the velocity of sales?
Zero demand months (no velocity, making forecasting difficult with a lot of zeros)
Abnormal demand. Where is it? (product type_region_warehouse_customer) When did it occur?

Variability

Which products impede velocity? Which increase the velocity of sales turned to cash?
Which products cause interruptions in manufacturing with the demand that is not constant
Where and what products exhibit lumpy demand? (regions_warehouses_strategic customers)
What are the seasonal products (product type_customer_region_warehouse)
What is our month’s supply of inventory of slow-moving products?
Are we stocked out of high volume/low variability products?

Variety

How do you manage the variety in the product offering? Is there a strategy?
How do you know the product mix is changing and when do you know it?
Do you know when to adapt your product to meet your plan for customization?
Do you manage the effect of variety (cost and customer service) during or after the fact?

If insight into these issues would provide information for which to make better decisions in managing your supply chain, consider a customized dashboard. After all, if you leave the software at the point when the last consultant left, how do you account for change?

Step 1: Experiment

You may want to download data from your software, or multiple software applications using a common data element. You might find sales and marketing exiting a product line and declaring it obsolete whilst manufacturing just entered a production run. If you lack visibility of forecasts and the master production schedule, this would be a good one to pursue.

Excel provides a great play tool to explore. Note the use of “play tool” and “explore”. Excel neither offers the scale or automation of advanced tools that deliver the dashboards. We’ll come back to that later.

Explore the data, twist and shout out how you need to see it for insight to action.

Step 2: Run the Data

This manual effort captures the results and may prove painful. These new views should drive action and provide insight for changes in forecasts, master production schedules, safety stock, and all the planning parameters such as lead time, transportation lanes, etc. Look for patterns and connections in the data.

Step 3: Analyze the Data then Show Results

Be prepared to dollarize your data. Know what variability could be reduced or what velocity could be increased, or what variety could be accommodated more quickly.

Consider a Six Sigma project to Define-Measure-Analyze and Improve. This structured and disciplined approach should lead to a Control Plan to keep the gains.

Step 4: Convince Management to Automate

Connecting to Step 1, now that you’re certain of what data you need, what time period, and where it comes from, it’s time to scope out the project to automate. Of course, if this comes in one person with supply chain depth of knowledge and product insight with current programming and coding skills, this step becomes much easier.

However, that’s not the world we live in.

Dollarize the value of the information and compare it to the cost to automate. The information may be relevant to the business for a long period of time, or the automation robust enough to allow for rapid changes made by key users to allow coders to move on to newer issues.

Step 5: Form a Partnership

Chances are a key user with data and product knowledge coupled with supply chain skills, aptitudes, and abilities that will form a bond with the coder who knows how to bring the data together to make it visible.

“You’ll know it when you see it” became the mantra of the quality revolution in the 80’s. Same here with graphs, charts, and tables in context with data all can agree upon.