It’s been a few months since I shared the learnings we’ve garnered from our Collaborative Flow Planning workshop – AKA – The Beer Game. Most supply chain professionals are familiar with this game and the supply chain planning pitfalls it exposes. Today, I’d like to focus on the second part of the game that was played, which was enabled by the Collaboration Workbench platform in JDA Flowcasting. We’re going to take a deeper look at how this platform was used to cut costs by two-thirds compared to those playing the game without the help of JDA Flowcasting.

Before we get started, here’s a quick recap of the three key takeaways we’ve learned from hosting our Beer Game workshops:

  1. The bullwhip effect is real and cannot be ignored. The four root causes (lack of visibility, siloed organizations/lack of communication, constraints like lot sizing/batches, and latency of information) will distort the demand signal unless they are addressed.
  2. Collaboration across any two nodes of the supply chain will yield benefits to both parties. You don’t have to solve everything all at one time.
  3. The level of trust and the value created between two collaborative parties has a direct correlation to each other.

Beer Game performance is measured by total network inventory costs ($0.40/case of network inventory) and customer service ($1.00/case/week back order cost). When Beer Game participants can only rely on their supply chain skills and expertise, their performance typically results in approximately $1,200-$1,500 in supply chain costs at the end of a 20-week game. Below is a common graphic showing how just a small blip of consumer demand cascades its way through our end-to-end, make-believe supply chain. It’s easy to see how demand increases as the planner’s role is further removed from the source of consumer demand.

Now, consider the results of the Beer Game with the help of JDA Flowcasting. With the exact same constraints used in the first game, performance of the participants in the JDA Flowcasting-enabled game yielded only $309 in supply chain costs!

What is driving such a big difference between these performance results? And how do we achieve this type of result every single time we play the game?

Live Supply Planning Collaboration & Analytics

JDA Flowcasting has long been known to enable end-to-end supply planning. This is based on a simple principle where you start by forecasting consumer demand and then calculate everything else across the supply chain. Developing a distribution requirements plan (DRP) like this has been around for a long time. What’s new about this?

To understand why this is different, first we need to take a moment to talk about the foundation of the JDA Flowcasting work platforms and how they differ from most of the traditional analytic tools used today. It’s important to note that JDA Flowcasting is not a reporting solution; it is a work platform that is integrated into the same data tables used by the demand and replenishment engines running the live supply chain. Now if the demand and replenishment technology being used is JDA Demand and JDA Fulfillment, then that linkage is a standard part of the JDA Flowcasting tool set. (And if you’re using different technologies, this integration can be built). The difference this combination of capabilities creates becomes obvious the moment you begin to leverage the power of these technologies.

When you “pull up” a JDA Flowcasting screen, dashboard or worksheet, you are literally looking at a live rendering of your current demand or replenishment plan, presented in a configurable format built to an end user’s perspective. This allows the user to review and analyze their live plan, and if they choose to do so, act. If something changes in that plan or if you change something in that rendering, you are changing the live plan.

A typical analysis usually involves pulling data from a combination of configuration tables and data warehouses. The analysis is conducted by staging and manipulating this data in an offline tool, most commonly Excel. The plan is then updated by translating the desired changes back into the live supply plan. We’ve all watched this; planners update the plan based on a screen print of a list of SKU numbers with numeric notes. The translation becomes tricky because the analysis was done on static data, in an offline tool, and during this time, the business did not stop. Implementing the changes can become a game of translating quantity changes in approximate time horizons to get directionally correct.

The JDA Flowcasting work platform changes that; a live rendering of your supply plan can be presented in a format configured to your business needs. The dashboard isn’t just a rendering of live data, but it brings powerful analytical tools that can be used to solve problems and prevent re-occurrence. When the analysis is complete and the planner wants to make changes, they are executed right in the dashboard, thus updating the JDA Demand and Fulfillment plan. While there are four JDA Flowcasting work platforms in total – Collaboration Workbench, Causal Analysis & Resolution, Prescriptive Scenario Analysis, and New Product Intelligence – we only use the Collaboration Workbench to execute the Beer Game.

The Flowcasting-Enabled Beer Game

The Collaboration Workbench was configured to represent the Beer Game network, following the exact same constraints used in the first game. Each user was provided a view specific to their role in the joint end-to-end live plan. For instance, if you were the factory role, the proper lead time was represented based on your position in the network.

As the game is played and consumer demand is realized at the retailer role each week, the resulting demand across the supply chain would be calculated through Collaboration Workbench. Inventory at each node was netted, and lead times and order rules were respected. The Collaboration Workbench in JDA Flowcasting did all the work, allowing each person along the supply chain to understand the impacts on their supply chain node.

In the Flowcasting-enabled Beer Game, each user simply needed to either agree or disagree with the recommendation the Collaboration Workbench was providing. If a person in the supply chain chose to disagree, any impacts to plan changes they made would have become immediately available to others in the joint supply chain. (No one has ever disagreed with a recommendation, by the way).

So how does this get us to a lowest cost/most efficient supply chain? The fact that people make changes and choices across the supply chain is nothing new. It happens every time a planner makes an override, a policy is changed, or a safety-stock target is altered, for instance. The problem, however, is that these decisions and changes are not well communicated and the impacts are not well known.

Collaboration Workbench solves this problem by creating a single platform everyone can work from to jointly manage the end-to-end supply chain. It breaks down communication barriers and the common causes of the bullwhip effect:

  • Siloed organizations
  • Lack of visibility
  • Latency of information

In the JDA Flowcasting-enabled game, the team acts as one, information is visible to everyone and immediately updated as the game progresses. It is the only difference between the two games, yet the results speak volumes to the value of addressing these common supply chain challenges.

Collaboration Workbench is being used in the industry today. It is being used to bridge the root causes of bullwhip between manufacturers, distributors and retailers. While we stack the Beer Game deck in our favor to enhance the concepts and workshop experience, there is nothing artificially stacked in our industry examples. We’ve seen network inventory decrease 20 percent while improving service. We’ve seen forecast accuracy realize double-digit improvement. All of this without an increase in the resources it takes to manage your day-to-day business.

You’ve probably noticed there was no mention about when supply chain constraints change. In the workshop, we intentionally keep the constraints the same between the two games. In future blog posts, I will explore how the other JDA Flowcasting work platforms – Causal Analysis & Resolution, Prescriptive Scenario Analysis, and New Product Intelligence – can help us understand alternative supply network strategies. I will also explore a common adoption approach and maturity curve as organizations grow into understanding how to leverage these newfound capabilities.

For additional information, watch this Flowcasting video.