As the speed and agility of supply chains evolves, the way in which labor resources are optimized also needs to evolve in order to keep pace with the changing buying behaviors of consumers. When retailers add the ability to change where and how products are fulfilled, they also need the ability to flex resources up or down to ensure they have the right number of people to support the real-time changes taking place at the store, in the distribution centers (DCs) and across their fleets.

The distribution workforce has historically been a full-time (often union), fixed-shift workforce that is viewed as a set annual cost. Retailers and their trading partner networks were primarily concerned with employee productivity as opposed to the number of employees needed. Schedules were established and shifts defined on an annual basis and unless there was a significant change to the distribution network, this number did not change with the exception of attrition. The problem with this approach is that a full-time fixed-shift workforce does not mirror the actual need for workers driven by the ever-changing demands placed on the DC.

The figure above represents throughput levels within a retail distribution facility. The red line represents throughput and the blue line represents typical staffing levels. The peaks and valleys created by too little or too much labor present the challenge to most employers. First, they must predict when the peaks and valleys will occur, then establish the appropriate staffing levels in an effort to minimize the use of temporary workers and overtime. Throughput is unpredictable, however, and is affected by a myriad of forces—economics, training, process, productivity, network change, order size, globalization, technology, demand volatility and SKU proliferation.

If you have too much labor, your total cost to serve will be excessive and you risk being either unprofitable or noncompetitive. If you have insufficient labor, you will either sacrifice service levels or lose valuable sales—neither of which is acceptable in an environment where retailers are fighting for customer loyalty.

In most DCs, labor makes up nearly 55% of total costs. If inventory levels are adjusted based on fluctuating demand, shouldn’t the variable costs associated with handling that inventory be adjusted as well? Workforce optimization requires a hierarchy of processes to ensure you have the right balance of full-time, temporary and overtime labor to handle the peaks and valleys in daily operations with the goal being to get the most work done at the lowest possible cost with the highest level of service. The hierarchy is based on a three-step process.

Step1: Strategic Workforce Forecast

The first step in optimizing your workforce is to forecast the projected throughput for the year. The inputs to the forecast generally include historical throughput, projected volume increases/decreases, employee productivity, new process improvements and network or facility changes. This exercise provides a forecast that plots the amount of labor required to perform the projected work by day, shift, job and zone. The outputs include an analysis of where there is too much labor and where there is too little, as well as the cost associated with that labor plan. The user can simulate the impact of adding overtime and/or temporary labor to fill the gaps and analyze the most effective balance of full-time, overtime and temporary resources to perform the projected workload.

Forecasting the projected workload and the labor requirements in advance provides visibility to where the projected bottlenecks may occur and enables the user to address issues well in advance.

Step 2: Workforce Schedule Optimization

The second step in aligning your workforce with actual demand is to schedule labor to optimally handle the inventory at the lowest possible cost. This is generally done on a monthly or even weekly basis to account for the most current demand information. Optimally scheduling your workforce requires taking into account individual employee productivity, labor rates and qualifications, and balancing that against that of other employees to determine the optimal time and activity for each individual. This analysis is more complicated than it may seem. In a typical warehouse with 100 employees and three shifts, the variables run into the millions. Only by leveraging advanced optimization technology can a company effectively calculate the optimal schedule.

The result is a schedule by employee, by day, by shift, by zone based on what is best for the overall facility.

Step 3: Milestone Workforce Optimization

The third step involves forecasting and scheduling labor around preplanned milestones that create a significant impact on demand such as new product introductions, promotions or seasonality. Unique spikes in demand created by event milestones are difficult to quantify. Looking at past history and massaging the data will provide insight into the potential impact and a basis for adjusting regular, overtime and temporary labor. It will also provide the basis for optimizing the scheduling of individual employees to ensure that the new and/or different work can be accomplished in the most cost-effective manner. The result is a labor resource allocation plan that is specifically tailored to the event.

Optimizing Labor to the Demand

Bottom line: Companies need to allocate resources in alignment with changing levels of throughput and inventory based on demand. Labor, as a key resource and cost within the distribution center, must be optimized to the demand. To ensure the proper level of labor, companies must address the following:

• Plan and forecast annual workforce budget.
• Establish appropriate level of regular workforce for projected work.
• Optimize the mix of Regular, Overtime and Temporary.
• Plan for seasonal changes, new product introductions and promotions.
• Continually optimize staffing levels by day, shift job and zone.

Peter Schnorbach is the Senior Director of Product Management for Manhattan Associates. Peter has responsibility for Labor Management and Slotting Optimization, and he has been instrumental in pushing labor management to the forefront as a mainstream supply chain solution. Before joining Manhattan he was the Vice President of Product Marketing at RedPrairie.  Prior to that, Peter spent 12 years in logistics positions at United Defense, Alliant Techsystems and Honeywell. Peter has a Law Degree from William Mitchell College of Law and a Bachelor of Arts Degree from the University of St. Thomas.