Performance Measurements and Metrics: An Analysis of Supplier Evaluation
Performance Measurements and Metrics
Based on the more current partnership approach, several techniques are used by companies to evaluate suppliers and measure performance. The first step in implementing any of the techniques being discussed is to determine the attributes that should be considered. A firm should focus on the attributes that it finds most important. Some attributes are easy to measure while others are not. A rule of thumb is to consider the total costs associated with a purchased product / service, not solely the purchase price. Some of the metrics that can be considered are (23):
- quality level
- service level
- correct quantity
- on-time delivery
- price/cost of product
- use of electronic data interchange (EDI)
- willingness to share sensitive information
- presence of certification or other documentation
- flexibility to respond to unexpected demand changes
- communication skills/systems (phone, fax, e-mail, Internet)
- quick response time in case of emergency, problem, or special request
- willingness to change their products and services to meet your changing needs
- willingness to participate in your firm’s new product development and value analysis
Some models are proficient in considering total costs, but they are usually very difficult to implement and time consuming. Thus, the resources available to the firm’s purchasing function will drive the firm’s model choice.
The three most common approaches are the categorical system, weighted-point average system and the cost-based system. The categorical system is the most subjective technique since it does not differentiate between the weights of the attributes considered. The weighted-point average system overcomes this drawback by assigning weights to each attribute. The cost-based system is the most objective of the three methods because it also considers non-performance costs.
In this method, the buyer chooses attributes that are most important to its particular situation. The buyer assigns either a preferred (+), unsatisfactory (-), or neutral (O) rating for each of the selected attributes to every vendor. Then the ratings are totaled for each vendor. For example, ratings resulting in scores of two preferred (++), one unsatisfactory (-), and one neutral (O) would total one positive (+). A comparison of total scores reveals the highest rated vendor. The representatives of the involved divisions agree upon the ratings (Hillman, 1993).
Categorical supplier measurement is the easiest method to implement but suffers from subjectivity. It does not provide a detailed insight into the supplier’s true performance because the attributes being measured are weighted equally. Also, because the representatives of the involved divisions assign the ratings, this method relies on an individual’s perception about performance and not on quantitative data.
In the weighted-point method, the relevant attributes are chosen and each are assigned a weight depending on the importance to the overall performance (24). The firm reaches a consensus on weight assignments to prevent or minimize subjectivity. The weight for each performance category is then multiplied by the performance score that is assigned to it. Finally, these products are totaled to determine a final rating for each supplier.
Firms often use the weighted point system because it is highly reliable and its implementation costs are moderate. In addition, it combines qualitative and quantitative performance factors into a common system. Because users can change the weights assigned to each performance category, or change the performance categories themselves depending on the strategic priorities of the firm, the system is flexible. The weighted-point method overcomes the subjectivity of the categorical system, but it has some drawbacks. It requires the buyer to specify the value of one performance measure relative to another, which is often difficult in practice.
Using the cost-based system, a buyer is able to quantify the additional costs incurred if a supplier fails to perform as expected. The total cost of doing business with the supplier can be calculated by the supplier performance index (SPI). This index is calculated for each item or commodity provided by the supplier and has a base value of 1. It is represented by the following formula: SPI = (Purchase Price + Nonperformance Cost) / (Purchase Price). As derived from the equation, the closer SPI is to 1, the better the supplier. Non-costs should include qualitative factors (14).
Benefits a buyer can achieve by using this approach include (26):
- the ability to source requirements based on total cost consideration
- a methodology to increase supplier accountability and control
- an equitable and consistent evaluation tool
- definition of supplier performance expectation
- communication of the firm’s buying priorities to suppliers
- the ability to perform sourcing risk assessment
- enhancement of internal communication for reporting critical supplier sourcing information
- the ability to provide positive supplier reinforcement
- a basis for a supplier award program
This system is the least subjective of the three because it quantifies the total cost of doing business by considering non-performance costs. The main difficulty in the use of the system is its complexity and its requirement that users have a developed cost accounting system. Although this sounds like an ideal way of dealing with costs, it is difficult to identify the costs of supplier non-performance.
Against their subjectivity and drawbacks, the categorical method, the weighted-point method and the cost-ratio method are the most widely used techniques in supplier evaluation due to their ease of use and implementation. Several other models and techniques have been proposed for supplier measurement and evaluation. These include: total cost of ownership approach, analytical hierarchy process ( Barbarosoglu and Yazgac, 1997), weighted linear model approaches (Lamberson et al. 1976; Timmerman 1986; Wind and Robinson 1968), mixed integer programming (Weber and Current 1993, discreet choice analysis experiments (Verma and Pullman 1998), matrix method (Gregory 1986), human judgment models (Patton 1996), interpretive structural modeling (Mandal and Deshmukh 1994), statistical analysis (Mummalaneni et al. 1996), linear programming models (Pan 1989; Turner 1988), and neural networks (Siying et al. 1997), multi-objective programming (Weber and Ellram 1993). Of these methods, Total Cost of Ownership (TCO) method stands out as a commonly endorsed approach.
The Total Cost of Ownership (TCO) approach is very similar to the cost ratio method. However, it is more comprehensive because it considers all the costs associated with quality, delivery and service. These costs include a number of non-value added activities such as service costs, receiving costs, quality costs (inspection, rework, reject costs), failure costs, and administrative costs including management time, maintenance, disposition and life-cycle costs. Lifecycle costs are costs incurred throughout the life of a product or service. These costs may include maintenance, downtime, repair, overhead, and idle time. In a study by Ellram, it was concluded that the price of a piece of production equipment for a firm’s operations was only 35 percent of the total cost of that piece of equipment over its life cycle. Despite the high percentage of the non-value added costs, firms tend to either underestimate or completely ignore them (7).
TCO is a proactive and comprehensive system. Rather than focusing only on price, it examines issues outside of the supplier’s cost structure. But using TCO presents the firm with new challenges. The method is complex to implement and maintain. Thus, it consumes a great deal of time.
Supplier evaluation methods are usually time consuming, so they are not performed frequently. In practice, many firms report supplier performance on a monthly or quarterly basis. In fact, some firms evaluate suppliers only once a year. However, frequent meetings with suppliers facilitate the prevention of inefficient practices at an early stage and encourage continuous improvement of suppliers. These assessments, however, are mutually beneficial only if both parties are willing to cooperate and provide the necessary inputs.
The information gained from supplier evaluation is valuable only when both buyer and supplier use it to improve their partnership. Supplier measurement systems are most commonly used for the following purposes: 1) Track performance of supplier; 2) identify supplier improvement opportunities; 3) develop supplier; 4) benchmark suppliers against best practices.