HOUSTON -- Kroger Co.'s success in reaping measurable benefits using labor-scheduling software for the front end is now prompting the chain to look at expanding the program to other areas of the store.
The labor-scheduling program at the front end has resulted in better customer service levels, improved employee relations, reduced production time in generating schedules and fewer errors, said Tom Murphy, vice president of management information systems at Kroger, Cincinnati.
"We measure our productivity in terms of hours per $100 of sales and we've seen some significant improvements in that," Murphy said.
Kroger is committed to providing customers with superior service at the checkout and matching the number of baggers needed to the customer count in specific timeframes. Implementing a labor-scheduling program has helped the chain make significant progress in these areas, he said.
The next step with labor scheduling involves exploring the possibility of expanding the program to additional departments, such as bakery.
However, implementing the program in other departments, where production issues may be much more complex, presents a greater challenge than at the front end, and that is one reason Kroger is still in the planning stages regarding expansion.
"Variable and fixed labor demand have to be considered," he said. "The variable tasks are based on service and production, while the fixed tasks, such as ordering and receiving, have to be identified and accounted for in the scheduling," Murphy said.
"We're experimenting with this right now and do not yet have all the answers," he added.
One way the chain is considering addressing these issues is by linking variable tasks to sales volume gleaned from point-of-sale data, Murphy said during a workshop session at the MarkeTechnics convention here last month sponsored by the Food Marketing Institute, Washington.
Labor scheduling for in-store bakery with a midnight to 6 a.m. doughnut production shift, for example, must take into account finished product inventory goals based on projected customer counts and number of employees needed to complete the task.
A store typically may experience a big rush on doughnuts from 6 a.m. to 9 a.m., during which time demand for customer service is at a peak. From 9 a.m. to noon, on the other hand, doughnut sales may slow considerably, but there is suddenly a need to clean up the back-room production area.
In bakery, the biggest challenge is figuring out the effect that special events will have on production requirements and thus labor-scheduling needs. Kroger is still working through these issues, he said.
Given those dynamics, retailers seeking to implement labor scheduling in many areas of the store, such as bakery, should launch a pilot program in a select group of stores before rolling out the program on a widescale basis, Murphy said.
As data is then gathered on hours needed for specific tasks and times of day, the schedule can be adjusted accordingly, he said.
During a question-and-answer period at the end of the workshop, Jay Stephens, director of operations research at Marsh Supermarkets, Indianapolis, asked panelists if, in their view, labor scheduling was still primarily focused on front-end activity or was truly ready for expansion into other departments.
"Labor scheduling is being used very successfully in perimeter departments by retailers who understand the production component, the service component and the blending of the two," said Kathy Springer, director of labor management at Senn-Delaney, a unit of Arthur Andersen, Los Angeles, who also took part in the workshop.
"These retailers have been able to identify the production component and understand the need to add additional hours for the service component. They also realign production so they have enough production completed to meet product demand and have the right amount of associates available to drive service," she said.
In a typical environment, the easiest way to implement labor scheduling beyond the front end is to figure out budgeted hours for the department and lay out production based on point-of-sale data and store standards, Springer explained.
The difference between those two is the service component. Retailers should then take a good look at production to determine how much can be moved into the day when the customer is there and tie that to service requirements as well, she added.
Stephens noted that Marsh has conducted more than 400 time studies in produce alone, which makes it difficult to blend that amount of data into a labor-scheduling package.
Springer said 400 time studies can be incorporated if there is a mechanism to manage the process and implement change as it comes up. "I find it's better to focus on a handful of key functions that occur in the area rather than so many, which may lead to diminishing returns," she said.