Lancaster University Management School Uses SAS® Software to Meet Product Sales Forecasting Challenge
Lancaster University Management School's (LUMS) Department of Management Science recently used SAS® software to meet a product sales forecasting challenge using data drawn from a well-known academic database, which contains data from a broad selection of stores and products owned by a leading US retailer. The database originated as a result of a partnership between the retailer and a leading school of business, and covers store-level scanner data collected over a period of more than seven years.
Lancaster University Management School (LUMS) is one of the UK's top research-led business schools. The Department of Management Science participates in various undergraduate, graduate and PhD degree programmes designed to provide students with logical, mathematical, project management, systems thinking, and information systems and technology skills.
In rolling out its masters courses, the department engages with industry partners that help provide students with academic and commercially focused work. SAS supports this approach through its academic programme. It does this by developing courses on which SAS® software is used, and by supporting the placement of students as interns at a range of clients, working on projects where they use SAS software to solve industry challenges.
The department has recently been using the latest in SAS® Analytics software to help students on a masters dissertation project address issues retailers in the fast-moving consumer goods (FMCG) sector face around inventory, stock management and product sales.
Specifically, the department is focused on helping retailers more accurately forecast product sales achieved by different stock keeping units (SKUs). FMCG retailers often use SKU forecasts to inform their inventory planning, reduce the incidence of products being either out of stock or overstocked, and in this way cut costs and increase profits.
This is the issue that the research team from the LUMS' Department of Management Science is looking to address, using the data in the prescribed database.
Deploying SAS® Software
The department used SAS® Retail Forecasting software to help create a demand forecast at the SKU level: maximising stock coverage while achieving more balanced inventory levels and reducing costs. In doing so, it helped provide a solution to many key challenges faced by the department's analytics team.
The first challenge is the data set, derived from the database itself. It is a large data set in terms of volume, consuming 19 gigabytes of storage, and includes information collected from 80 food stores over an eight-year period, encompassing 29 broad product categories and a total of 116 individual products. Using the SAS software, in conjunction with different econometric modelling techniques, the department's team set out to forecast the sales performance of each of the products – an exhaustive process that would be almost impossible to carry out manually.
It made extensive use of a range of SAS procedures to analyse data in SAS data sets; to produce statistics, tables, reports, charts, and plots; to create SQL queries; and to perform other analysis and operations on the database.
The department first used the experimental GLMSELECT procedure, a recent addition to SAS/STAT® software, to pinpoint which external variables are likely to be most important to forecasting product sales and to take these into the model. It then used a range of SAS procedures including RATE, MODEL, FORECAST, FACTOR and VECTOR to develop the models, segment the data set into seven different product categories and evaluate the performance of the models.
The SAS® Retail Modelling software was used in this highly complex project. As Dr Tao Huang, formerly Lancaster University Management School's chief researcher on the project (now at Imperial College London), stated: "First, it is able to handle recursive modelling and second, SAS is particularly strong in handling large data sets."
Professor Robert Fildes added: "In addition to the ability to handle large databases, the SAS software enabled us to put together sophisticated procedures that both handle the data and develop models for it. The level of functionality that SAS supports has enabled us to successfully overcome the challenges we were facing at the outset of the project."
The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.
Copyright © SAS Institute Inc. All Rights Reserved.
Lancaster University (Forecasting)
Help FMCG retailers accurately forecast product sales achieved by different SKUs. Help to inform inventory planning, reduce incidence of products being out of stock or overstocked, cut costs of the process and increase profits.
SAS® Retail Forecasting software helps create a demand forecast at the SKU level: maximising stock coverage while achieving more balanced inventory levels and reducing costs.
SAS® software ideally suited to this project. Demonstrates capabilities in advanced retail modelling. Able to handle recursive modelling and manage large data sets. Delivers procedures that both handle and develop models for the data.
“There were two key differentiators that made SAS essential for this project. First, it is able to handle recursive modelling. And second, SAS is particularly strong in handling large data sets.”
Dr Tao Huang
Academic researcher, Centre for Forecasting, Lancaster University Management School