Attempting to Answer the Wrong Question
Traditional MRP (material requirements planning) systems fail again and again. As Simon Eagle reminds us, writing in Demand-Driven Supply Chain Management:
[D]riving replenishment execution through materials requirements planning (MRP)-dependent demand network with today's high levels of forecast inaccuracy inevitably leads to unbalanced inventories that cause supply chain and production instability, or variability, as schedules frequently have to be amended to prevent service issues. This leads to the development of excessive inventories, excessive lead times and necessitates the use of unplanned capacity. However, adoption of the Demand-Driven Supply Chain Management (SCM) approach, especially in 'make to stock' supply chains, allows planned service levels to be achieved from half the average inventories, with far higher overall equipment effectiveness (OEE) and significantly shorter lead times. 
If you work in supply chain management, inventory management, production planning or purchasing, you’ve experienced the reality of this day after day.
The reason driving supply chain execution off MRP and forecasts fails, I believe, is because it attempts to answer the wrong question.
Forecast-driven MRP attempts to tell you precisely what to build or buy, and when. It is always precise and it is virtually always precisely wrong.
Is There a Better Answer?
Writing in the outstanding book The Theory That Would Not Die, Sharon Bertsch McGrayne offers this tidbit regarding Bayesian thinking:
Bayesians could… combine information from different sources, treat observables as random variables, and assign probabilities to all of them, whether they formed a bell-shaped curve or some other shape. Bayesians used all their available data because each fact could change the answer by a small amount. Frequency-based statisticians threw up their hands when Savage [Jimmie Savage] inquired whimsically, "Does whiskey do more harm than good in the treatment of snake bite?" Bayesians grinned and retorted, "Whiskey probably does more harm than good." 
NOTE TO READER: If you’d like to have a gentle, non-technical introduction to Bayes’ Theorem, watch this YouTube video from Veratasium.
Keep reading. I not trying to convert you to a statistician or mathematician--really!
Combining information from different sources
There are some key factors to be noted in the statement above. Bayesian thinking allows us to better understand our world, our situation, and make proper adaptations incrementally as new information is made available. It does this by combining relevant information of different types, from different sources, and then guiding incremental adjustments to our thinking.
This really works!
Real-world applications of Bayesian thinking cracked the “unbreakable” German Enigma code in World War II, helped us hunt down Russian submarines during the Cold War, and guided the calculations that provided a sound basis for starting the U.S. Workers’ Compensation Insurance program.
Answering the Wrong Question the Wrong Way
Going back to the Jimmie Savage question in the passage above: Traditional MRP approaches try to use statistics to answer the question:
“Does whiskey do more harm than good in the treatment of snake bit?”
The answer from this approach must be binary. It’s “yes” or “no.” Traditional MRP wants to tell you to act—or not act—based on forecasts and statistical analysis.
Answering the Right Question the Right Way
Demand-driven MRP (DDMRP) as promulgated by the Demand Driven Institute shows us a better way.
Strategically placed and sized buffers are built and maintained by combining relevant information from different sources and treating observables as variables in the maintenance of the buffers themselves. Here is a summary of the kinds of data that are combined into the day-by-day maintenance of buffer sizes and statuses:
- Average Daily Usage (ADU)
- Demand Variability Factors
- Replenishment Lead Time Factors
- Planned Adjustment Factors (based of foreseeable future events)
- Demand Spikes within a predetermined Spike Horizon
- Minimum Order Quantity (where applicable)
Combining these relevant data into the calculation of the buffer size and its current status, DDMRP provides the answer to the right question the right way.
Instead of trying to answer in a binary way, DDMRP can tell you:
“Whiskey probably does more harm than good in the treatment of snake bite.”
More directly, a glance at the status of any buffer tells everyone the answer to the truly critical question:
“How effectively is this buffer—at this moment—protecting FLOW in my supply chain and what actions, in what priority, will probably protect FLOW in my supply chain most effectively?”
Based on this bit of data, and looking at this bit of data across all the buffers in the supply chain, managers can rapid make accurate and timely decisions about execution priorities and quantities.
As Robert E. Kass, a Bayesian at Carnegie Mellon University says,
"Bayes Theorem…. says there is a simple and elegant way to combine current information with prior experience in order to state how much is known…. It makes full use of available information, and it produces decisions having the least possible error rate." [Emphasis added.] 
Decisions with the Least Possible Error Rate
Wouldn’t you like to start making supply chain execution decisions that have the highest likelihood of being right, and with the right priorities to protect FLOW (read: profit)?
Then we suggest that you become truly demand-driven (which does not mean, make-to-order, by the way), and we can help. Leave your comments below or feel free to contact us directly, if you prefer.
 McGrayne, Sharon Bertsch. The Theory That Would Not Die: How Bayes' Rule Cracked the Enigma Code, Hunted Down Russian Submarines, and Emerged Triumphant from Two Centuries of Controversy. New Haven: Yale University Press, 2012. – Referencing Erickson W.A., ed. (1981) The Writings of Leonard Jimmie Savage: A Memorial Selection. American Statistical Association and Institute of Mathematical Statistics.
 Eagle, Simon. Demand-Driven Supply Chain Management: Transformational Performance Improvement. New York: Kogan Page, 2017.
 McGrayne, ibid.