Exam objective 2.6 asks you to describe supply chain planning and demand forecasting in Dynamics 365 Supply Chain Management. These capabilities solve the fundamental business problem of having the right amount of stock, in the right place, at the right time, without wasting money on excess inventory. For the MB-920 exam, you need to understand what these tools do, why businesses use them, and how they differ from simply reacting to orders as they come in.
Jump to a section
A simple way to picture Supply Chain Planning and Demand Forecasting
A kite festival supply stand sells colourful kites, string, and paper rolls every spring. The stand's owner, Mei, must decide how many kites to order from her supplier three months before the festival starts. Her stand has limited shelf space and she cannot return unsold kites. If she orders too many, she loses money on leftover stock. If she orders too few, she misses sales and angers disappointed children. This is the core challenge of supply chain planning: matching supply with uncertain demand. Mei uses her past sales records, the weather forecast for festival day, and the number of school groups that have booked visits to predict demand. She also builds a small buffer stock of popular designs for latecomers. This buffer is like safety stock in a real supply chain. She then plans her ordering schedule, delivery timing, and staffing levels around this forecast. Mei's process maps exactly to how Microsoft Dynamics 365 Supply Chain Management handles master planning and demand forecasting. She starts with historical data, adds external signals, runs a forecast, checks inventory, and generates a replenishment plan. The only difference is that Mei uses a notebook and a hunch, while a planner uses Dynamics 365 to automate the calculations, handle thousands of products, and update the plan in real time as new information arrives.
Supply chain planning and demand forecasting are two closely related processes that help a business decide what to buy, make, and move. They replace the old way of operating, which was to wait for a customer to place an order and then scramble to fulfil it. That reactive approach leads to long delivery times, angry customers, and either stockouts or piles of unsold goods. Dynamics 365 Supply Chain Management provides a set of tools that plan proactively.
First, let us define demand forecasting. A forecast is a prediction of future customer demand for a product over a specific period, such as the next week, month, or quarter. The system generates this forecast by analysing historical sales data, seasonal patterns, and external factors like promotions or holidays. For example, a company selling umbrellas might see a sharp spike in demand every November. The forecast captures that pattern and predicts how many umbrellas will be needed next November. The user can also manually adjust the forecast if they know something the data does not show, such as a planned marketing campaign. The forecast is not a single number but a range with a probability attached. Dynamics 365 uses a feature called Demand Forecasting to run statistical models and produce these predictions. It can also pull in data from outside the system, such as weather reports or economic indicators, to improve accuracy.
Once the demand forecast exists, the system uses it as input for supply chain planning. The core tool for this is called Master Planning. Master planning answers two questions: What do we need to buy or make, and when do we need it? It takes the forecast and also considers current inventory levels, open purchase orders, existing sales orders, and lead times from suppliers. Lead time is the number of days it takes for a supplier to deliver an item after you order it. The planning engine then calculates net requirements. For instance, if the forecast says you will sell 100 laptops next month, you already have 20 in stock, and 30 are coming from a supplier next week, then your net requirement is 50 laptops. The system then creates planned orders to purchase or manufacture those 50 laptops. It also schedules them to arrive before the predicted demand date, taking lead times into account. This process is called supply planning or replenishment planning.
There are two main strategies that planners use within master planning, and the MB-920 exam tests your understanding of both. The first is make-to-stock (MTS). In this strategy, products are manufactured or purchased based on a forecast, without waiting for a specific customer order. This works well for standard products with stable demand, like bottled water or office paper. The second is make-to-order (MTO). Here, products are only made or bought after a customer places a firm order. This is used for custom or expensive items where holding finished goods inventory is too risky, such as a bespoke wedding dress or a large industrial machine. Dynamics 365 supports both strategies, and the system can mix them for different items in the same company.
A critical concept to understand is the planning horizon. This is how far into the future the master planning run looks. A typical planning horizon might be 12 months. The system does not generate planned orders for the entire 12 months in one go. Instead, it generates them for a shorter period called the frozen zone, then re-plans regularly as new data arrives. The planning frequency can be daily, weekly, or monthly depending on the business. After master planning runs, a supply planner reviews the suggested planned orders. They can firm them, which means converting a planned order into a confirmed purchase order or production order. They can also reschedule, cancel, or change quantities. This human review is called the planning cycle. Dynamics 365 also supports constrained planning, where the system respects known capacity limits, such as a machine that can only produce 500 units per day. If the plan would exceed that capacity, the system flags a capacity constraint or suggests overtime.
Finally, the output of supply chain planning feeds directly into procurement, manufacturing, and logistics. Purchasing agents see planned purchase orders and start negotiating with suppliers. Production managers see planned production orders and schedule factory workers. Logistics teams see plans for shipping and receiving. All of these departments work from a single, unified plan, which prevents the chaos of each team planning in isolation. The system also supports what-if analysis. A planner can change a forecast or a lead time in a simulation copy of the data to see how the plan would change, without affecting the real plan. This is a powerful feature for making decisions about inventory investment or supplier performance.
Collect Historical Data
The system gathers past sales data, usually for two or more years, broken down by product, location, and time period. This data forms the foundation of the demand forecast.
Run Demand Forecasting
Dynamics 365 applies statistical models to the historical data to predict future demand. The user can add external factors like promotions or weather data. The output is a forecasted demand number for each product and period.
Run Master Planning
The master planning engine takes the demand forecast plus current inventory, open orders, and lead times. It calculates net requirements and generates planned purchase orders and planned production orders to cover the forecasted demand.
Review and Firm Planned Orders
A human supply planner reviews the generated planned orders. They can firm them (convert to real orders), reschedule them, adjust quantities, or cancel them. This step prevents the system from acting on faulty assumptions.
Monitor and Adjust the Plan
After orders are firmed, the planner monitors actual sales against the forecast. If actual demand diverges significantly, they rerun the forecast, adjust safety stock levels, or initiate what-if analysis to explore corrections without affecting live data.
An IT professional working in a manufacturing or distribution company does not just turn on Dynamics 365 and let it run. They must configure, validate, and maintain the supply chain planning and demand forecasting system. Here is a step-by-step walkthrough of a realistic scenario.
Sarah is the supply chain manager at a company that produces organic protein bars. She uses Dynamics 365 Supply Chain Management. The company has 50 different product flavours and sells through both retail stores and its own website. Sarah's job is to ensure that the right bars are on the shelf when customers want them, without wasting ingredients that expire.
First Sarah runs the demand forecasting process. She opens the Demand Forecasting workspace. The system has been collecting two years of historical sales data, broken down by flavour and by month. The forecasting engine automatically detects seasonal patterns: chocolate flavour peaks in winter, fruit bars peak in summer. She reviews the generated forecast for the next six months. She notices that the forecast predicts a 20% increase in demand for peanut butter bars, based on a new marketing campaign the sales team entered into the system. Sarah manually adjusts the forecast for one flavour that the algorithm underestimated because of a recent recipe change. She then publishes the final demand forecast. This forecast becomes the official prediction that the master planning engine will use.
Next, Sarah runs master planning. She navigates to Master Planning and selects the plan for her distribution centre. She configures the planning run to cover 12 months but to only generate firm planned orders for the first 3 months. The rest are tentative signals to suppliers. She clicks Run. The system takes about 10 minutes to calculate. When it finishes, the system shows a list of planned orders. There are planned purchase orders for raw ingredients like oats, protein powder, and dates. There are also planned production orders for each flavour of bar. Sarah reviews the list. She sees that the system suggests buying 5,000 kg of oats, but the supplier's lead time is 14 days. The forecast expects a demand spike in 10 days. This is a problem. The planned orders will arrive too late. Sarah uses the what-if analysis feature. She creates a simulation that adds a second supplier with a 7-day lead time. The system recalculates and shows that with this change, the plan would work. Sarah then contacts the supplier to expedite the first order and starts onboarding a second supplier. She firms the first few planned orders so the purchasing team can act on them today.
Sarah also sets up safety stock levels. Safety stock is extra inventory kept as a buffer against demand variability or supply delays. She configures Dynamics 365 to calculate safety stock automatically based on the forecast error. For high-demand flavours like chocolate, safety stock is set to two weeks of average demand. For low-demand flavours like kiwi, safety stock is set to four weeks because the ingredients are harder to source. The system will trigger a reorder as soon as inventory dips below the safety stock level plus anticipated demand during lead time. This is called a reorder point. Sarah reviews these settings monthly.
Finally, Sarah runs a weekly supply planning meeting using Dynamics 365 dashboards. She uses Power BI visuals embedded in the system to show the gap between forecast and actual sales. She also reviews the planned order coverage report, which highlights which items are understocked or overstocked. Her team uses this data to decide which planned orders to firm, which to delay, and whether to run promotions on overstocked flavours. The IT team supports all of this by ensuring the data integrations from the online store and the warehouse management system are running smoothly. They also manage user permissions so that only planners can adjust forecasts and firm orders.
This scenario shows that the IT professional's role is not just to install the software. It is to configure the planning parameters, interpret the outputs, validate the logic, and enable the business users to make informed decisions. The exam tests whether you understand this workflow, not the obscure technical settings.
The MB-920 exam focuses on the concepts, not the configuration screens. You do not need to memorise navigation paths. You do need to understand what each term means and how the pieces fit together. Here is exactly what you need to know.
First, the exam tests the difference between forecasting and planning. A demand forecast is a prediction. Master planning uses that prediction to create a supply plan. A common trick question asks: 'Which process generates planned purchase orders?' The correct answer is master planning, not demand forecasting. Demand forecasting only generates predictions; it does not create orders. Another trap rephrases 'master planning' as 'supply planning' or 'materials planning'. These are synonyms. Know all three terms.
Second, the exam tests the difference between make-to-stock (MTS) and make-to-order (MTO). You must recognise which strategy fits which scenario. For example, a question might describe a company that makes standard light bulbs and ask which planning strategy they use. The answer is make-to-stock. A different question might describe a custom furniture maker; the answer is make-to-order. The exam also tests a hybrid called configure-to-order (CTO), where a base product is stocked but customised after the order. Know all three by name and use case.
Third, the exam loves to test safety stock and lead time. A question might ask: 'If lead time increases from 5 days to 10 days, what happens to the reorder point?' The correct answer is that the reorder point increases because you need to order earlier to cover the longer wait. Another question might ask which factors the system uses to calculate safety stock. The correct answer includes demand variability and supplier reliability. Know that safety stock is a buffer against uncertainty.
Fourth, the exam tests the planning horizon and planning cycle. A question might ask: 'What is the purpose of the planning horizon?' The answer is to define how far into the future the system should calculate requirements. A trap question might suggest that the planning horizon is the same for all items. That is false. Different items can have different horizons. For example, a high-value custom item might have a longer horizon than a fast-moving consumer item. Know that the horizon is configurable.
Fifth, the exam tests what-if analysis. A question might describe a planner who wants to see the impact of a supplier delay without affecting live data. The correct answer is to run a simulation or what-if analysis within a copy of the plan. A trap is to suggest that the planner should change the live forecast and rerun. That would corrupt the official plan. Know that what-if analysis is a separate, safe environment.
Finally, the exam tests the role of the demand forecast in the broader system. It feeds into both master planning and also into other modules. For example, the forecast can drive budgeting, capacity planning, and resource planning. A question might link the forecast to the budgeting process. Know that it is a shared dataset, not just for supply chain. Be careful with questions that confuse the output of demand forecasting (a table of numbers) with the output of master planning (planned orders). They are completely different.
Demand forecasting predicts future customer demand using historical data and external signals, while master planning generates planned purchase and production orders based on that forecast.
Master planning calculates net requirements by subtracting current inventory and open orders from forecasted demand, then schedules replenishment respecting supplier lead times.
Make-to-stock (MTS) produces goods based on a forecast before any customer order is received, while make-to-order (MTO) produces goods only after a customer order is confirmed.
Safety stock is a buffer inventory held to protect against variability in demand or supply, and it is calculated based on forecast error and lead time uncertainty.
The planning horizon defines how far into the future the master planning engine looks, and it can be set differently for different items or product groups.
What-if analysis lets a supply planner simulate changes to forecasts or lead times in a safe copy of the plan without affecting the live operational data.
The demand forecast is a shared dataset that feeds not only master planning but also budgeting, capacity planning, and resource planning across the organisation.
These come up on the exam all the time. Here's how to tell them apart.
Demand Forecasting
Predicts future customer demand using historical data and statistical models
Output is a forecasted demand number for each product and time period
Does not create purchase orders or production orders
Master Planning
Takes the demand forecast as input and calculates what to replenish
Output is planned purchase orders and planned production orders
Creates actionable supply suggestions for the planner to review
Make-to-Stock (MTS)
Production is based on a forecast, not a specific customer order
Products are finished and held in inventory before sale
Best for standard, high-volume products with stable demand
Make-to-Order (MTO)
Production is triggered by a confirmed customer order
Products are not finished or held in inventory until ordered
Best for custom, expensive, or low-volume products
Safety Stock
Extra inventory held as a buffer against demand or supply uncertainty
Calculated based on forecast error and lead time variability
Not expected to be used in normal operations, only in unexpected situations
Cycle Stock
The inventory that is expected to be sold during the lead time period
Calculated as average demand per day multiplied by lead time in days
Represents the normal working inventory that rotates regularly
Mistake
Demand forecasting and master planning are the same thing.
Correct
Demand forecasting predicts what customers will want. Master planning uses that prediction to decide what to buy or make. They are two separate steps in a sequence.
The terms sound similar and both involve planning, so beginners assume they are interchangeable. The exam explicitly tests this distinction.
Mistake
Master planning runs automatically without any human involvement.
Correct
Master planning generates suggestions, but a human planner must review them and convert them into firm orders. The system does not place purchase orders by itself.
Marketing materials often talk about 'automated planning' which makes people think no human is needed. In reality, human judgement is critical for handling exceptions.
Mistake
Safety stock is the only inventory you need to hold.
Correct
Safety stock is just a buffer. You also need cycle stock (the inventory you expect to sell during lead time) and possibly seasonal stock. Safety stock sits on top of these.
The word 'stock' is ambiguous. Beginners think safety stock covers everything, but it only covers unexpected variability.
Mistake
Forecasts are always accurate if you have enough historical data.
Correct
Forecasts are always wrong to some degree. They become more accurate with better data and models, but they remain estimates. Planning must account for forecast error.
People assume that more data equals perfect prediction. In reality, unexpected events like a pandemic or a competitor's promotion break any forecast.
Mistake
What-if analysis changes the actual production plan in the system.
Correct
What-if analysis runs on a simulation copy of the data. It does not change the live plan unless the user explicitly applies the changes. It is a 'try before you commit' tool.
The word 'analysis' suggests a passive read, but beginners worry it will alter real data. The exam wants you to know it is a safe sandbox.
Mistake
Make-to-stock is always better than make-to-order.
Correct
Each strategy suits different products. Make-to-stock is good for standard items with stable demand. Make-to-order is better for custom or expensive items where holding stock is risky. Neither is universally superior.
Beginners often assume one method is 'best' because they have only seen one in their own experience. The exam tests that you recognise when to use each.
Reveal each answer, then mark whether you got it right. Score 60%+ to unlock the next chapter.
Demand forecasting predicts future customer demand using historical data and external factors. Master planning uses that forecast to create planned purchase and production orders. The forecast is the input; the plan is the output.
No. Master planning creates planned orders that a human planner must review and firm. The system does not place purchase orders or start production runs without a person confirming them.
Safety stock is extra inventory kept as a buffer against unexpected spikes in demand or delays from suppliers. It ensures you can still fulfil customer orders when things do not go as planned.
Make-to-stock produces goods based on a forecast before any customer order arrives. Make-to-order only produces or purchases goods after a customer order is received. Choose make-to-stock for standard products; choose make-to-order for custom or expensive products.
The planning horizon is the amount of time into the future that the master planning engine examines when calculating requirements. It is configurable per item or plan, and it determines how far ahead the system generates planned orders.
Yes. Dynamics 365 supports what-if analysis, which lets you create a simulation copy of your plan. You can change the forecast or lead times and see the impact on planned orders without changing the live data.
You've finished Supply Chain Planning and Demand Forecasting. Continue through the MB-920 study guide to build a complete picture of the exam.
Done with this chapter?