Executive Briefings

Supply Chain Didn't Quite Fit After Apparel Company Changed Its Business Model

When Oxford Industries shifted from being a traditional manufacturer of men's clothing to a company focused on the design and marketing of products made by others, it needed new technology for demand and supply planning.

Oxford Industries began life in 1942 as a domestic manufacturer of basic, button-down shirts for mid-level retailers, particularly department stores. In recent years, however, the company has shifted its business model to one focused on apparel design and marketing, with third-party producers handling manufacturing.

As part of this transformation, the Atlanta-based company embraced a brand-focused business strategy. In 2003, Oxford acquired the island-inspired Tommy Bahama operations, followed by the 2004 acquisition of Ben Sherman-a well-known London-based brand made famous by the popularity of its shirts among British rock stars.

Oxford's legacy business units, Lanier Clothes and Oxford Apparel, also evolved. As one of the leading suppliers of men's tailored clothing to retailers, Lanier Clothes designs and markets suits, sports coats, suit separates and dress slacks. While continuing to sell these under private labels, it also has licensed a number of well-known brands, including Geoffrey Beene,  Kenneth Cole and Dockers. These products span a wide price range and are sold at national chains, department stores, specialty stores and discount retailers throughout the United States. Oxford Apparel's products range from dress shirts and western wear to suit separates and golf apparel, designed mostly for private-label customers like Lands' End, Federated Department Stores and Men's Wearhouse. Oxford Industries also sells through 55 of its own stores.

As the company's business changed, so did the demands and complexities of its supply chain. Oxford already faced the demand and supply challenges inherent in the fashion industry: multiple, short seasons and hard to predict variables of color, size and style. Outsourcing manufacturing to Asia and other parts of the globe, thereby extending the length of the supply chain, added another level of complexity.

In the late 1980s, early in its transformation process and prior to the acquisition of Tommy Bahama and Ben Sherman, Oxford realized that it needed to bring its business divisions up to speed with more robust information technology. After completing the implementation of a company-wide enterprise resource planning system, the company contracted with an independent consulting firm to determine where it should invest time and money to further increase operational efficiencies and performance. The result of that in-depth study ultimately led to Oxford Industries' decision to implement two solutions from JDA Software: Demand Planning and Master Planning.

"The JDA Demand solution offered us the opportunity to significantly improve our forecast accuracy," says John Baumgartner, chief information officer at Oxford Industries.

With so many possible permutations of size, style and color for each of its products, improving forecast accuracy was critical. Prior to implementing JDA Demand, Oxford relied on its retail customers' demand forecasts for its private-label products, as well as information provided by the company's own sales associates. If too much or too little product was created based on the retailer's or the sales associates' forecast, both Oxford Industries and that customer paid the price via lost sales or markdowns.

JDA Demand enabled the company to better understand consumers' evolving requirements and current trends, along with historical buying patterns, resulting in the ability to create more accurate forecasts and synchronize demand for replenished product with sources of supply. Oxford Industries can now compare its forecasts with those of its retail customers to ensure that the right amount of product is manufactured, leading to improved collaboration and service levels with its trading partners.

Baumgartner gives an example in which a key customer's seasonal demand forecast for a particular style was considerably higher than the forecast Oxford generated using JDA Demand. During one of the companies' regular collaboration meetings, the retailer agreed to use a middle-of-the-road forecast at the start of the season and to monitor actual results in order to determine which forecast was more reliable.

"In that situation, our forecast to produce less product turned out to be more accurate, and the retailer agreed to use our forecast for the remainder of the season," Baumgartner says. "Without our ability to offer a statistically sound forecast, we would have manufactured at least 10 percent to 15 percent more product, forcing the retailer to eventually mark it down."

In addition to the direct impact on that customer, there also was a collateral impact across the entire organization, says Baumgartner. "Since production capacity is finite, producing those unnecessary units would have meant that some other order would have been late or short of meeting that customer's needs. Using JDA Demand better positions us to collaborate with our customers, enabling us to make the right amount of product at the right time, and that benefits everyone."

Being able to aggregate and granulate demand forecasts as needed is another important aspect of the JDA Demand solution, says Danny Halim, vice president of supply and manufacturing solutions at JDA, Scottsdale, Ariz. "Demand planning is not just about figuring out how much of a product will sell, but taking the style/color/size complexity and aggregating that up so you can look at demand by product line or geographic region or distribution channel," he says. "At the same time, you need to be able to look in the other direction and see demand in a very granular way, down to the store cluster level."

Every level of an organization uses demand information, Halim adds. "Whether you are an executive looking at rolled-up demand or a buyer looking at very specific detail, the numbers should all match up. Tools like JDA Demand improve demand accuracy because of the intelligence and algorithms it employs and the business processes that it enables."

For Oxford Industries, having an accurate forecast also is critical to ensuring that it purchases the right amount of raw material. Oxford buys many of its goods on an order-by-order basis from offshore third-party producers. But, due to manufacturing complexities, some of the company's product is acquired on a "cut-make-trim" basis. In these cases, Oxford supplies some or all of the raw materials and contracts with third-parties to cut, sew and finish the apparel product, or it manufactures the product in its own factories.

"Typically, companies in the apparel industry have to look six to nine months out for pre-season planning," says Halim. "If they are not able to do that, they can get into a position where they may have to source product from vendors with a shorter lead time at a much more expensive rate. Our product gives them the visibility they need to plan ahead."

To better align supply with demand and to manage production among its suppliers, Oxford also implemented JDA Master Planning. At the time it acquired this solution, Oxford was sourcing goods through a combination of owned and contract factories, with much of the product serving to replenish its customers' inventories as needed. For all of Oxford Industries' owned factories and for many of its contract facilities, manufacturing capacity was a vital consideration. Supply chain planning needed to consider not only basic plant capacity but also what is known in the apparel industry as "sub-capacity" or the ability to deal with products requiring matched plaids or having other features that call for specialized sewing skills and equipment. Additionally, production planning had to take into consideration raw-material availability, as well as manufacturing and customer lead-times.

Moreover, because of the nature of Oxford Industries' product lines, the ever present complexities of size, color and style had an added layer. Size measurements typically had two dimensions. Dress shirts, for example, had separate neck and sleeve measurements. Similarly, tailored clothing had chest and coat-length sizes and slacks had waist and inseam dimensions.

"For just one pant style with four different color options, we might be managing a combined SKU count of more than 100," Baumgartner says. "Manually, the best we could do was plan at the color and style level using only total manufacturing capacities."

Once a broad plan based on those parameters was complete, the company would attempt to determine if all of the other constraints could be satisfied, including raw-material availability and manufacturing sub-capacities. According to Baumgartner, the company's planners would spend Monday through Thursday each week compiling the plans. Any issues that were identified throughout the process would have to be resolved on Friday for the following week.

The implementation of JDA Master Planning leveraged the solutions' automated functionality to compile product information and production constraints to generate weekly sourcing and inventory plans from style to the SKU level. The solution also simultaneously considered factory capacities including special features, raw-material availability, and manufacturing and customer lead-times. Since Master Planning generated a first version of the supply plan by noon each Monday, Oxford Industries' planners had four and a half days to resolve any issues to accommodate unplanned demand, which translated to an 85-percent improvement in planning efficiency.

"With Master Planning, we have the capability to plan at the style, color, size and dimension level-something that was not possible to do manually," Baumgartner explained. "The solution really changed the supply planning process within our legacy businesses from a planning activity to truly managing the critical issues."

Although the company's sourcing model has since shifted from a typical manufacturing process to more of a purchase process, manufacturing and customer lead-times, SKU-level decisions and some capacity constraints still need to be factored into the supply planning process, he says. "Master Planning provides the tools to let managers manage instead of serving as data-entry technicians."

"Businesses that are good at demand planning will be more effective at removing risk from the equation and raising the probability of higher sales and profits," adds Baumgartner. "In terms of supply planning, a company that can master the management of SKUs across the sea of supply chain constraints will optimize its inventory management processes and increase customer satisfaction."

RESOURCE LINKS:
Oxford Industries, www.oxfordinc.om
JDA Software, www.jda.com

Oxford Industries began life in 1942 as a domestic manufacturer of basic, button-down shirts for mid-level retailers, particularly department stores. In recent years, however, the company has shifted its business model to one focused on apparel design and marketing, with third-party producers handling manufacturing.

As part of this transformation, the Atlanta-based company embraced a brand-focused business strategy. In 2003, Oxford acquired the island-inspired Tommy Bahama operations, followed by the 2004 acquisition of Ben Sherman-a well-known London-based brand made famous by the popularity of its shirts among British rock stars.

Oxford's legacy business units, Lanier Clothes and Oxford Apparel, also evolved. As one of the leading suppliers of men's tailored clothing to retailers, Lanier Clothes designs and markets suits, sports coats, suit separates and dress slacks. While continuing to sell these under private labels, it also has licensed a number of well-known brands, including Geoffrey Beene,  Kenneth Cole and Dockers. These products span a wide price range and are sold at national chains, department stores, specialty stores and discount retailers throughout the United States. Oxford Apparel's products range from dress shirts and western wear to suit separates and golf apparel, designed mostly for private-label customers like Lands' End, Federated Department Stores and Men's Wearhouse. Oxford Industries also sells through 55 of its own stores.

As the company's business changed, so did the demands and complexities of its supply chain. Oxford already faced the demand and supply challenges inherent in the fashion industry: multiple, short seasons and hard to predict variables of color, size and style. Outsourcing manufacturing to Asia and other parts of the globe, thereby extending the length of the supply chain, added another level of complexity.

In the late 1980s, early in its transformation process and prior to the acquisition of Tommy Bahama and Ben Sherman, Oxford realized that it needed to bring its business divisions up to speed with more robust information technology. After completing the implementation of a company-wide enterprise resource planning system, the company contracted with an independent consulting firm to determine where it should invest time and money to further increase operational efficiencies and performance. The result of that in-depth study ultimately led to Oxford Industries' decision to implement two solutions from JDA Software: Demand Planning and Master Planning.

"The JDA Demand solution offered us the opportunity to significantly improve our forecast accuracy," says John Baumgartner, chief information officer at Oxford Industries.

With so many possible permutations of size, style and color for each of its products, improving forecast accuracy was critical. Prior to implementing JDA Demand, Oxford relied on its retail customers' demand forecasts for its private-label products, as well as information provided by the company's own sales associates. If too much or too little product was created based on the retailer's or the sales associates' forecast, both Oxford Industries and that customer paid the price via lost sales or markdowns.

JDA Demand enabled the company to better understand consumers' evolving requirements and current trends, along with historical buying patterns, resulting in the ability to create more accurate forecasts and synchronize demand for replenished product with sources of supply. Oxford Industries can now compare its forecasts with those of its retail customers to ensure that the right amount of product is manufactured, leading to improved collaboration and service levels with its trading partners.

Baumgartner gives an example in which a key customer's seasonal demand forecast for a particular style was considerably higher than the forecast Oxford generated using JDA Demand. During one of the companies' regular collaboration meetings, the retailer agreed to use a middle-of-the-road forecast at the start of the season and to monitor actual results in order to determine which forecast was more reliable.

"In that situation, our forecast to produce less product turned out to be more accurate, and the retailer agreed to use our forecast for the remainder of the season," Baumgartner says. "Without our ability to offer a statistically sound forecast, we would have manufactured at least 10 percent to 15 percent more product, forcing the retailer to eventually mark it down."

In addition to the direct impact on that customer, there also was a collateral impact across the entire organization, says Baumgartner. "Since production capacity is finite, producing those unnecessary units would have meant that some other order would have been late or short of meeting that customer's needs. Using JDA Demand better positions us to collaborate with our customers, enabling us to make the right amount of product at the right time, and that benefits everyone."

Being able to aggregate and granulate demand forecasts as needed is another important aspect of the JDA Demand solution, says Danny Halim, vice president of supply and manufacturing solutions at JDA, Scottsdale, Ariz. "Demand planning is not just about figuring out how much of a product will sell, but taking the style/color/size complexity and aggregating that up so you can look at demand by product line or geographic region or distribution channel," he says. "At the same time, you need to be able to look in the other direction and see demand in a very granular way, down to the store cluster level."

Every level of an organization uses demand information, Halim adds. "Whether you are an executive looking at rolled-up demand or a buyer looking at very specific detail, the numbers should all match up. Tools like JDA Demand improve demand accuracy because of the intelligence and algorithms it employs and the business processes that it enables."

For Oxford Industries, having an accurate forecast also is critical to ensuring that it purchases the right amount of raw material. Oxford buys many of its goods on an order-by-order basis from offshore third-party producers. But, due to manufacturing complexities, some of the company's product is acquired on a "cut-make-trim" basis. In these cases, Oxford supplies some or all of the raw materials and contracts with third-parties to cut, sew and finish the apparel product, or it manufactures the product in its own factories.

"Typically, companies in the apparel industry have to look six to nine months out for pre-season planning," says Halim. "If they are not able to do that, they can get into a position where they may have to source product from vendors with a shorter lead time at a much more expensive rate. Our product gives them the visibility they need to plan ahead."

To better align supply with demand and to manage production among its suppliers, Oxford also implemented JDA Master Planning. At the time it acquired this solution, Oxford was sourcing goods through a combination of owned and contract factories, with much of the product serving to replenish its customers' inventories as needed. For all of Oxford Industries' owned factories and for many of its contract facilities, manufacturing capacity was a vital consideration. Supply chain planning needed to consider not only basic plant capacity but also what is known in the apparel industry as "sub-capacity" or the ability to deal with products requiring matched plaids or having other features that call for specialized sewing skills and equipment. Additionally, production planning had to take into consideration raw-material availability, as well as manufacturing and customer lead-times.

Moreover, because of the nature of Oxford Industries' product lines, the ever present complexities of size, color and style had an added layer. Size measurements typically had two dimensions. Dress shirts, for example, had separate neck and sleeve measurements. Similarly, tailored clothing had chest and coat-length sizes and slacks had waist and inseam dimensions.

"For just one pant style with four different color options, we might be managing a combined SKU count of more than 100," Baumgartner says. "Manually, the best we could do was plan at the color and style level using only total manufacturing capacities."

Once a broad plan based on those parameters was complete, the company would attempt to determine if all of the other constraints could be satisfied, including raw-material availability and manufacturing sub-capacities. According to Baumgartner, the company's planners would spend Monday through Thursday each week compiling the plans. Any issues that were identified throughout the process would have to be resolved on Friday for the following week.

The implementation of JDA Master Planning leveraged the solutions' automated functionality to compile product information and production constraints to generate weekly sourcing and inventory plans from style to the SKU level. The solution also simultaneously considered factory capacities including special features, raw-material availability, and manufacturing and customer lead-times. Since Master Planning generated a first version of the supply plan by noon each Monday, Oxford Industries' planners had four and a half days to resolve any issues to accommodate unplanned demand, which translated to an 85-percent improvement in planning efficiency.

"With Master Planning, we have the capability to plan at the style, color, size and dimension level-something that was not possible to do manually," Baumgartner explained. "The solution really changed the supply planning process within our legacy businesses from a planning activity to truly managing the critical issues."

Although the company's sourcing model has since shifted from a typical manufacturing process to more of a purchase process, manufacturing and customer lead-times, SKU-level decisions and some capacity constraints still need to be factored into the supply planning process, he says. "Master Planning provides the tools to let managers manage instead of serving as data-entry technicians."

"Businesses that are good at demand planning will be more effective at removing risk from the equation and raising the probability of higher sales and profits," adds Baumgartner. "In terms of supply planning, a company that can master the management of SKUs across the sea of supply chain constraints will optimize its inventory management processes and increase customer satisfaction."

RESOURCE LINKS:
Oxford Industries, www.oxfordinc.om
JDA Software, www.jda.com