This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
Does anybody have experience with demand sensing overall and have ideas on how to model that in Alteryx?
It is about sensing/predicting when an order is about to come in a short-term horizon (next weeks as time horizon). There would be historical and external factors information available. Granularity is very high using location-sku level. The focus is on logistics/inventory positioning.
Let me know if somebody has done something like that and/or you have an idea of which elements of Alteryx to use in order to model demand sensing!
This varies widely company to company and the data sources that they have available, but in general terms, you will want to see what downstream data that you have access to in order to clean, harmonize, and enrich based on the data architecture and road map that you are working towards. The data from your retailers must be compatible with your ERP system data, or transformed in order to fit the standardized data model. You must also ensure that the downstream data is at the same level of reporting granularity, and therefore keep things in mind like the frequency in which you gather the data to ensure your reporting and modeling has the synchronization of the two systems in consideration. It is also important to think about the objective of the analysis: more volume vs greater market share, when to launch new product, reaction to upswings in demand to capture additional revenue, free up cash flow, stabilizing production schedules, reducing transportation costs, etc and what models are most appropriate.
The actual math involved will of course vary depending on what you are attempting to accomplish, but taking the steps outlined above should be a good start forward in your journey!