How to define your demand planning process and its key steps.

Defining the demand planning process is crucial for any business seeking to maximize profits. From the supply chain perspective, demand planning plays a key role in contributing to the company's objectives. In this regard, let's explore some key aspects of how the supply chain relates to demand forecasting.

Demand Planning

The function of defining the demand planning process is to ensure the availability of products at the time required by the customer. At the same time, it should aim to eliminate any costs in the manufacturing process that do not add real value to the product.

For our demand planning to be effective, the information it is based on must be as accurate, manageable, qualitative, and quantitative as possible. This way, a reliable and appropriate sales forecast can be established for the products we sell.

Its management is the first step in the supply chain, from which the rest of the processes should be structured and adjusted.

How much and when do I need to manufacture?

What level of inventory should I maintain?

How do I organize my work shifts?

Do I have the necessary raw materials?

Benefits for Your Organization

By defining the demand planning process, we can respond to the rest of the supply chain processes with greater precision, making our operations as efficient as possible and obtaining numerous benefits for our organization, such as reducing obsolescence, increasing service levels, and optimizing inventory.

Maximize your profit: By improving your service level, you will reduce lost sales, generating a positive impact on your bottom line.

Reduce your inventory: Reliable forecasting will allow you to adjust your stock levels, increase your cash flow, and reduce balance sheet inventory.

Streamline your processes: You will have greater visibility into the other processes in the supply chain, such as production planning and procurement.

Demand Planning Cycle

To carry out the demand planning cycle, a statistical forecast will be made using historical sales data and other business knowledge as a basis.

While this is a global definition of the process, to achieve success, structured steps must be followed to obtain a reliable, measurable, and common sales forecast for the entire organization.

The existence of the demand planner role is crucial to ensure the proper execution of the process. They will be the ones to articulate and govern the process, involving different areas of the organization.

The following are the key critical aspects of each step in the demand planning process:

1. Load Historical Data

First, the historical sales data should be loaded into the tool that supports the demand planning process in the organization.

It is important to define the data source because, depending on it, the information will need to be treated differently. Generally, two types of historical sales data are used:

Sales Orders: These define actual customer orders, and the orders associated with real demand should be identified, removing any duplicate information.

Shipments: These define the demand fulfilled for customers, and it needs to be corrected with the level of service. There may be cases where not everything the customer requests leaves the warehouse, which could distort the sales forecast.

2. Data Cleansing

On the other hand, key suppliers in the affected areas must be identified. Efforts should be made to maintain contact with them and, if necessary, obtain alternative sources of supply. Accelerating the recovery time of a partner not only helps survive the threat but also ensures their own future success.

To launch the statistical forecast, only recurring customer demand will be considered. The data loaded into the system should be analyzed to clean it of any irregularities that may affect the forecast. For example, removing promotional sales or unusual demand behaviors.

3. Statistical Forecasting

Once the sales data has been corrected, we will be ready to generate the statistical forecast. There are many models available, ranging from simple ones like moving averages to more complex ones like first and second-order exponential smoothing.

Depending on the nature of the demand, different forecast parameters should be selected. Based on the behavior, we will define the planning level: item, product family, or business dimension. Additionally, we will define the level of temporal aggregation based on the required accuracy.

4. Business Intelligence

There are certain business and market aspects that cannot be captured by statistical models and must be taken into account to establish the final demand plan.

For this reason, the sales department should review the statistical forecast to update any "extraordinary" sales that are planned, such as new product launches, promotions, or campaigns.

A clear example is the business intelligence acquired during the recent pandemic (impact of the Coronavirus on sales forecasting), where each business has gained insights and, to the extent possible, will be better equipped to react to possible future outbreaks.

5. Demand Plan

Once a demand plan is defined, consensus must be reached among all the areas of the organization involved. The objective is to ensure that the defined forecast can be met by the other processes involved in the supply chain, both from a service and cost perspective.

Can I manufacture everything that has been set in the demand plan?

What is the impact on inventory and production costs?

Do I have the necessary raw materials for production?

Once validated by all parties, it will be distributed throughout the organization.

6. Continuous Improvement: "What gets measured gets improved"

Finally, the demand planning cycle must be executed under a continuous improvement model that is repeated periodically, allowing for information analysis to evaluate the performance of the process.

Some of the most important indicators through which we can measure the accuracy of our forecasts include:

DFA (Demand Forecast Accuracy): Measures the reliability of the forecast, that is, how close the forecast is to the actual sales value. This indicator will always be positive, with a maximum value of 100%.

BIAS: Measures the deviation between the forecast and actual sales, with a positive or negative sign. A BIAS of 0% would indicate that our forecasted data is exactly equal to our sales.

If you want to learn how we can help you reduce costs and optimize your planning process with Anlytics Demand Planning, contact us and choose the day and time that suits you best!