Building efficient product portfolios at John Deere and Company.(OR PRACTICE)

John Deere & Company (Deere), one of the world's leading producers of machinery, manufactures products composed of various features, within which a customer may select one of a number of possible options. On any given Deere product line, there may be tens of thousands of combinations of options (configurations) that are feasible. Maintaining such a large number of configurations inflates overhead costs; consequently, Deere wishes to reduce the number of configurations from their product lines without upsetting customers or sacrificing profits. In this paper, we provide a detailed explanation of the marketing and operational methodology used, and tools built, to evaluate the potential for streamlining two product lines at Deere. We illustrate our work with computational results from Deere, highlighting important customer behavior characteristics that impact product line diversity. For the two very different studied product lines, a potential increase in profit from 8% to 18% has been identified, possible through reducing the number of configurations by 20% to 50% from present levels, while maintaining the current high customer service levels. Based on our analysis and the insights it generated, Deere recently implemented a new product line strategy. We briefly detail this strategy, which has thus far increased profits by tens of millions of dollars.

Subject classifications: industries: machinery; marketing: retail/product line optimization; production: applications.

Area of review: OR Practice.

1. Introduction

Deere & Company (Deere) manufactures equipment for construction, commercial, and consumer applications, as well as engines and power train components. As a major player in many equipment markets, Deere maintains multiple product lines; within each line, there may be several thousand, to several million, different product variants. Variants are built by selecting for each feature available on a machine (e.g., engine type, transmission, and axle) one of a number of possible options (e.g., 200, 250, or 300 horsepower (HP) for engines). Not all options are compatible; a feasible combination of options is called a configuration.

Deere speculates that maintaining too many configurations reduces profits, by elevating what Deere calls complexity cost. This cost, over and above the inventory carrying costs of each configuration, captures factors such as reduced manufacturing efficiency, frequent line changeovers, and the general overhead of maintaining documentation and support for a configuration. This definition is similar to that given in Thonemann and Brandeau (2000, p. 1), where complexity cost is "the cost of indirect functions at a company and its suppliers that are caused by component variety; complexity cost includes, for instance, the cost of designing, testing, and documenting a component variant." In this paper, we describe the marketing and operational methodology and tools we developed to reduce Deere's complexity costs by concentrating product line configurations while maintaining high customer service, thus elevating overall profits. We illustrate our work with applications to two lines at Deere; details of the products have been disguised, but the lines differ in significant ways (e.g., costs, profits, and sales), making them a diverse test bed for our optimization algorithm.

A primary component in our algorithm is our customer migration model, quantifying the behavior of Deere's customers: A customer may want a specific configuration, but if his or her first choice is unavailable, he or she may migrate to an alternative configuration that does not differ too greatly from the first choice. Using actual sales, along with customer segmentations and part-worths utilities provided by Deere, we probabilistically model every customer as individually identifying a set of acceptable configurations, sorted in decreasing order of preference: their migration list. When the top configuration on his/her list is not available, a customer will buy the next available configuration. When no configuration is available, the customer defects to a competitor.