Forecasting vs Customers Orders?

The idea behind the videos and published articles is to share ideas so we can all become more informed about what is happening in the global supply Chain world. I am doing a series of live videos every Tuesday at 4PM CAT (2PM GMT) discussing various aspects of supply chain.

This article supports a live video of the same title screened on Tuesday 21 January 2020 and discusses three 2 points on Forecasts and the resulting problems caused within organizations. Point 3 discusses a 21century solution using DDMRP and basing supplier orders on customer demand.

To watch the video on our UTube channel, you can follow this link

  1. Our reliance on forecasts and the resulting problems

In today’s complex environment, we have ever increasing supplier lead times because of globalization. Most businesses will import some percentage of their requirements from an International source. In the last 2 decades, China and India have become the major sources of raw and packaging material, either in printed or unprinted form. This has added months to our supplier lead times.

At the other end of the value chain our customer tolerance times are shrinking because there are so many alternatives available to them.

This is where the dilemma starts.

We need to forecast much earlier so that we can place orders on suppliers to get our materials on time for conversion. So, in an environment that has a 3-month supplier lead time, and a 2-month production cycle, we are trying to forecast what the market will demand 5 – 6 months’ into the future

DDMRP-variability-bullwhip-3-e1488289541502Meetings (S&OP) between Sales and Marketing, the Supply chain and Finance start out amicably, but tend to end up as a finger pointing exercises because of the difference between forecast and actual demand and the low customer service level that results. The Master Scheduler is continually having to revise production plans, and also revise Supplier orders, leading to continual revision and cancellation of orders.

In the factory, the schedulers chop and change the production schedule based either on a shortage of materials, or on stocked out critical Finished product.

Internal and external relationships become strained, reinforcing the Silo culture. Within the Supply Chain, production is screaming because of the disruptions, QA is upset because congestion and continual change overs lead to quality defects and the factory staff become demoralised because they’ve worked overtime to release critical product which then gets stuck in quarantine. Our Suppliers start threatening to increase pricing or they become non responsive to our pleas to either cancel or expedite orders.

  1. The dilemma of forecast accuracy

Depending on the industry, forecasts are based on various inputs.

Historical sales are manipulated and extrapolated into the future. Added to this, we have input on Customer promotional deals and on internal promotions to drive sales and new product launches or discontinuations. These numbers are then manipulated statistically and revised to meet the annual budget so that annual sales targets can be achieved.

Cynical I know, But the point is, no matter how much effort you put in to get an accurate forecast, they are always inherently wrong.

This is starting point for the master scheduler who, using MRP now needs to accurately predict what materials (and their precise quantities) are needed to satisfy a customer order based on due date that is 5 months in the future.

The impact of variability determines that accuracy & reliability can only get worse over time for dependent items.

  1. The 21st century solution – Using Customer Orders to drive Supplier Orders.

DDMRP gives us the ability to determine materials requirements based on Customer Orders. Irrespective of the level of accuracy of the forecast, we can now ensure an excellent customer service level, and at the same time stock only the materials we are converting and the finished goods the customer is buying. This is achieved by decoupling the lead times, correctly sizing our buffers and placing them to support the decouples lead times. Replenishing is now based on buffer levels and not by Customer Due date.

DDSCN 2The major impact of decoupling is the reduction of variability. The decoupled lead times can be more easily managed to avoid variability, and the buffers then absorb any remaining variability, such as order spikes.

The strategic buffers also allow us to hold the correct materials, intermediate and finished goods stock to satisfy demand and keep our working capital at a minimum because they are held at the lowest cost possible.

In short, the buffers supply down stream demand and then signal to the next upstream buffer when they need to be replenished. This cycle works from customer back to the Supplier, creating the ultimate pull system.

Should you have any questions or comments, please in the comments section or email to You can also call me on +27 82 777 0922.

Thanks to the Demand driven institute for creating such an awesome methodology that is changing the face of supply chains around the world.

About the author

Dave Hudson is a supply chain and operations specialist and executive coach with over 30 years’ experience, and currently 1 of 5 endorsed Demand Driven instructors on the African continent.