What is Demand Planning?
Demand planning is a supply chain management process of forecasting, or predicting, the demand for products to make sure that they will be delivered and satisfy customers. The goal is to strike a balance between having sufficient inventory levels to satisfy customer needs without having a surplus.
Importance/Characteristics of Demand Planning
The market can shift on a dime and demand plans need to move at the speed of the changing market. If demand plans can’t be adjusted with agility, companies could end up with stock-outs and unhappy customers, warehouses full of unused inventory, unhappy finance managers, and millions of dollars in wasted capital.
In a perfect world, demand planners must stay before the market rather than merely reacting thereto, and make decisions supported near real-time market data, instead of solely on historical data. That’s not always possible, but with the arrival of cloud-based planning platforms, it’s closer to reality than ever before.
Elements/Steps in the Demand Planning Process
Demand planning may be a complex process that typically includes the subsequent elements:
- Data collection from internal and external sources on the factors known to predict or influence demand
- Statistical analysis of sales, inventory, and other data.
- Modeling the info to predict future demand
- Collaboration with suppliers, manufacturers, salespeople, and other stakeholders to gather information on events that might affect demand, like promotions and production delays.
Our Methodology for Improving it
The Key steps in our process model are:
- Assess the demand forecast needs of the down-stream customer
- Identify roles and responsibilities of the organization on the demand side including Sales, Marketing, Product Development, and Category Management
- Map historical data, clean and harmonize the data for Statistical modeling
- Recommend appropriate demand forecast models and help tune statistical models to arrive at unbiased baseline forecasts. See more details on model tuning.
- Design the sub-process for Sales Forecasting including promotional planning and integrate it with the CRM process in the organization. Event Modeling can be for repeatable promotional events and holiday spikes but can also be used for modeling Black Swan events. See the blog entry on black swans.
- Design a consensus process and define appropriate review reports.
- Define the rules for forecast reconciliation and demand consensus
- Define exception management for demand and supply volatility. Our focus is to enable you to manage the process by exception. We have data mining capabilities to dig into your process and help you develop exception management
- Define the process for identifying supply constraints and reverse feedback. Demand Management is a continuous process and thrives on constant communication, feedback, and action. We help you implement processes that are self-sustaining and continuously improving
- Define holistic metrics and processes for continuous improvement. We help you design end-to-end Value Chain Metrics including key measurements for the demand chain. It is important to measure your entire demand chain not just the demand planners.
Future of Demand Planning
We are currently testing the Azure Machine Learning algorithms for demand planning to see what automated ML can bring to the planning community.
One thing that amazed me – was how fancy the idea of machine learning can get. It uses new terminology for classic concepts in statistical modeling – Anomaly detection, Training the Model, Testing the Model, etc. Regardless, a good time-series forecast should be part of the planner’s toolkit for developing a baseline forecast – a good starting point for the entire demand consensus process within the S&OP.
Demand Time-series can be divided into three periods:
The model initialization phase is the period to compute a pre-estimation averaging. Most efficient engines may use the entire training phase to compute this average and then re-use this for training the model.
The testing phase is the out-of-sample phase which is a rigorous examination of the quality of the model that is considered the best fit of the patterns in the data. Typically, this is the last few periods of the time series.
Once the algorithm passes this final examination, then it is used to develop the forecast.
Click here to get in touch with us to have more insights on demand planning.