Data-driven workforce optimization is fueled with accurate forecasts
Aug 19, 2020 • 3 minWorkforce optimization is the process of creating staffing schedules that not only comply with employment contracts and regulations but also match with the forecast-driven workload and employee skills. For retailers, workforce optimization helps ensure that every shift is filled with not only an adequate number of people to avoid over- or under-staffing but also the right people based on skillset, availability, and legal requirements.
Retail Workloads Consist of Fixed and Volume-Based Tasks
The first step in workforce optimization is to identify workload based on demand levels. The workload typically consists of two types of tasks: fixed and volume based.
Fixed tasks recur on a consistent basis and have the same (or close to the same) workforce requirements every time. Such tasks, including administrative work and daily store opening duties, can typically be set once in a workforce system and then referenced as needed during schedule creation.
However, approximately 80% of a retail store’s workload is made up of volume-based tasks—tasks that change from day to day and week to week based on customer footfall, inbound delivery volumes, and online orders. Such tasks can be impacted by recurring variations in demand patterns, like seasonality; business decisions, like promotions; and external factors, like weather. These complexities make machine learning essential for accurate workload projection.
Forecasts That Give Insight into the Workload of Volume-Based Tasks
Volume-based tasks include running cash registers, stocking shelves, supporting in-store customers, unloading deliveries, and picking online orders. For further insight into these types of tasks and the workload associated with them, a workforce optimization system must have access to different types of forecasts, including:
- Customer footfall forecasts at 15-minute increments. These granular forecasts are used to identify workload requirements for tasks, such as customer service and cashier duties, that correlate with how busy a store is at any given moment and that must happen at specific times.
- Inbound daily delivery forecasts.These forecasts provide a detailed schedule on a daily cadence of which items will be delivered and when, providing valuable insight into workload needs related to inventory management tasks, such as unloading and shelving. The inbound volumes usually follow a very different pattern than customer footfall or sales, which is why it is important to consider them separately. Unlike tasks that are connected to footfall, there is usually some flexibility with the timing of inventory-related tasks.
- Online channel forecasts. For omnichannel retailers, online channel forecasts deliver information that can help determine the workload level and timing for any tasks related explicitly to ecommerce, including picking and sorting. Online orders typically follow a very different cadence from in-store shopping, so a workforce optimization system needs separate forecasts to determine the shift and task patterns of staff responsible for managing online-related tasks and to calculate whether additional staff may be required.
Optimized Schedules Must Combine Fixed and Volume-Based Task Information with Employee Data
The volume-based task information provided by all three forecast types must be combined with fixed task information to accurately predict the total workload level at any day or time. A workforce optimization system should be able to take into account footfall-dependent, fixed, and flexible tasks and optimize them together so that they will not interfere with each other or with customer needs, e.g., access to aisles, during peak busy times.
Once the different types of forecasts and fixed tasks are optimized together, the system can take that workload and cross-reference it with employee profiles (which contain details about contracts, legislation, skills, and individual needs) to create schedules that meet both business and human requirements. This vast amount of data is nearly impossible to manage manually, making it crucial to harness the power of intelligent algorithms.
Retailers who introduce a workforce optimization system can expect significant business benefits. For example, Swedish grocer Coop Värmland saw personnel costs drop by between 6% and 8% as a result of implementing such a system. In addition, employees who feel their schedules take their needs into account tend to be to be happier and more satisfied with their employer. Overall, workforce optimization is likely to lead to a reduction in unnecessary costs and an improvement in service levels.
To realize the full potential of workforce optimization, the system must have access to highly accurate forecasts that cover customer footfall, daily deliveries, and online channel information. This data, when automatically combined with detailed employee profiles, allows the system to algorithmically develop high-quality schedules that match with the workload and require minimal manual intervention, enabling store teams to focus on other critical business needs and tasks.