Durability is a quality dimension which I believe most people have a similar view of, but which is hard to formally define and therefore can be defined in many different ways.
O’Connor (2001) defines durability as an item’s ability to “withstand wear out mechanisms, such as fatigue, wear, etc”. For non-repairable products this view gives a simple and convenient approach.
When looking at repairable products, the situation becomes more complex. Garvin (1984), defines durability for repairable products as “…the amount of use one gets from a product before it breaks down and replacement is regarded as preferable to continued repair.” This view implies that the durability is to some extent based on decisions made by the user. The decision whether to repair a product or to buy a new one would, Garvin elaborates, be based on such factors as repair costs, personal valuations of time and inconvenience, downtime losses and relative prices. Taking the decisions made by the user into consideration, as Garvin does, is in my opinion very relevant. We must be aware that the period during which a product will be used is limited not only by technical aspects.
I find it necessary to use two different properties, one which I refer to as durability and one which I call economic life. I define economic life to facilitate my analysis of external factors that affect the choice regarding when to replace trucks. For practical reasons, I use the Volvo definition of durability, a definition from 2004 that has been agreed upon by Volvo 3P, Volvo Powertrain, Volvo Truck Corporation and Renault Trucks. This definition is specific for trucks and would need to be adjusted to be applicable also for other products.
24
Durability is the products ability to perform its required function under stated conditions of use, and under stated conditions of maintenance, until the failure of a component, which has a significant cost compared to the expected residual truck value, and which needs to be overhauled12 or replaced.
Economic life is the period during which it is considered attractive, according to the owner’s business conditions, to have the product in use.
The customer I refer to is the one who pays for the usage of the truck, whether it is a purchase or a lease. For a private customer the attractiveness can be decided by such factors as comfort, fuel consumption, if the customer feels proud of the truck etc. For commercial customers however, who comprise the lion’s share of the truck market, the truck is a tool used for providing services that generate revenue. For them the economic life will be decided largely by the ability of the product to provide service at a profitability level that is competitive with the main alternative which is buying a newer truck. When it is considered smarter to replace it, the product most often still has a value, at least a selvage value, but the economic life is considered to be over.
The difference between the durability and economic life, in my view, is that durability defines how much use one can get until the cost of getting the truck running is substantial compared to the residual value, while economic life is how much use the an owner gets before he considers it to be better to replace it. Taking the decision to replace it does not necessarily require that it needs to be repaired. For instance, if a product becomes technologically obsolete, because there is a rapid technological development of new products, the product does not become less durable but its economic life might be shortened. Likewise, if you need to pay some fee to use an old product (for instant road tolls), that product does not become less durable, but its economic life is reduced. This view of these attributes implies that durability is largely technical. Economic life on the other hand is unique for each market, segment etc, while they all have different conditions. A product like a truck usually have
12 Here “overhaul” refers to a major repair.
25
several users and owners and both needs and wants of the owners and users might vary greatly between different countries, conditions of use etc. When one user decides that it is not profitable to keep using the truck, there is often another user with other needs or a different business model that is willing to buy the truck. I therefore choose to view economic life as a property based on the circumstances each customer act within. Many potential limitations of economic life such as regulatory environment and needs of reliability are different between markets and segments, but often relatively homogenous within them. Thus, when the economic life is over in one market or segment, it may continue in another.
For a product like a truck, built up by a large number of components, it is a matter of interpretation what makes the truck to be “worn out”. If some small and cheap components fail, they can probably be replaced and the truck must not necessarily be taken out of use. If we only need to make small repairs to keep the truck in service for a long time, the truck can be considered as quite durable. However, if we make big repairs, we need to start thinking about whether it really is the same truck as before. If we always change the components that break without any limitations, the truck will of course always be able to provide service.
In order to measure durability of trucks we usually measure something like the “technical life” of the truck. We define some unit that we measure it in, for trucks the number of kilometres travelled seems most appropriate. We also need to define clearly what repairs we allow before we consider the “technical life” to be over.
26
Figure 5: An example of a lifetime percentile curve
At Volvo Powertrain the durability of for instance an engine type is generally measured as a mileage at which a certain percentage of the engines in some population have failed. The measure can be applied also on a component level and it is stated as for instance L10 = 800 000 km, meaning that 10% of the units will fail before 800 000 km. Currently there are 12 components for the engine that are defined as “durability components”. The components are based on the criteria that they should either demand a so called major overhaul or have a repair cost exceeding a certain amount. When at least one of these components fails, there is a durability failure and the current mileage is our durability measurement. While also a large number of small failures together can create high total repair costs for the customer and thereby cause the customer to sell the truck earlier than planned, there is currently a discussion in the company about having a durability definition that takes total repair cost into consideration.
2.9.1 Classic durability economics theory
Much of the early research that was done on economical aspects of quality was concerned with product durability and its impact on the profitability of manufacturing companies.
However, most of it focused on the question of what durability a manufacturer should have in order to maximise profits in a monopolists market (see for instance Kleinman and Ophir, 1966, Swan, 1970).
27
Some models have also been made to explore optimal durability in a perfect market.
Kleinman and Ophir (1966) conclude that an increased interest rate will decrease optimal durability but that the demand of service from the products has no effect on it. These results are though somewhat shadowed by the fact that Swan (1970) proved some assumptions and thereby conclusions in the article to be wrong.
A general feature of the research on durable goods theory is that it so far consists of numerous models, whose assumptions are often arguable and which limit the validity and applicability of the conclusions severely. One of these assumptions (employed in Swan, 1970 and Sieper and Swan, 1973 and criticised by Avinger, 1981 and Waldman, 2003) is that units of service are perfect substitutes in consumption irrespective of the age or durability of the product supplying the services. One can easily see how there could be a difference between the service supplied by a brand new truck and that of an old one. Driving comfort, safety, reliability and fuel consumption are only a few important aspects that could be age-dependent.
Planned obsolescence
One way to create planned obsolescence is to frequently introduce changes in style that increases the difference between new and used trucks (Waldman, 2003). With a bigger difference it is easier to increase the price of new units without having the old ones as competition that is limiting the price.
Adverse selection
The concept of adverse selection applied on the automobile second hand market was described in Akerlof (1970). It is founded on the assumption that there will always be an asymmetry in information when a used product is sold, meaning that the seller has more knowledge about the quality of the individual car than the buyer does. If we assume that the buyer only knows something about the group of cars defined by model, production year etc, this is what will define the price. The consequence is that the sellers that own cars with lower quality than the group average will have the most benefit from selling. The owners of cars with over average quality on the other hand have little incentive to sell while they probably will not be able to get a price reflecting the high quality. The essence of this market mechanism is thus that the majority of the second hand market turnover will consist of cars of relatively low quality and the asymmetric information will overall reduce the activity of the
28
second hand market. The adverse selection, it should be noted, is not as applicable when there are warranties defining a minimum quality level. After the warranty period however, the mechanism is applicable as long as the buyer can not have as much knowledge as the seller about the individual item for sale.
If adverse selection is highly present, then the second hand market will be constrained, leading to a lower pricing of used cars (Hendel and Lizzeri, 1999). A lower second hand price and lower second hand market turnover will naturally effect sales of new goods, but the effect on the new car price from adverse selection is not absolutely clear. The lowered second hand price affects the new good price negatively, while it increases the cost of ownership for the buyer of the new car. On the other hand, a lower quality of the second hand market makes the used cars less substitutable for new ones which allows for higher pricing of new cars (Hendel and Lizzeri, 1999).
2.9.2 Potential implications of durability on business
The most relevant previous work that I have found on this topic is an article by Saleh et al (2006). In his article, Saleh discusses the issue of optimal durability from three different perspectives: the customer, the manufacturer and the society at large. For each interest party he lists a number of things that might be affected by the durability of a system and whether he considers it to be mainly positive or negative for that party. Saleh uses the case of a spacecraft as an example but the ideas are meant to be applicable in a more general sense. I have studied each of the proposed potential consequences of a change of durability and judged whether it may be relevant also for the case of heavy-duty trucks. While most of the consequences of extending the lifetime are reverse when shortening it, I am assigning consequences only to extending the design lifetime. In my view, these potential consequences depend on the actual durability rather than the design lifetime, but the expected value of the durability is the same as the design lifetime.
Potential implications to an extended design lifetime proposed by Saleh that I judge as relevant for trucks can be found in Table.
29
Table 2: Potential consequences of an extended design lifetime (A=Advantage, D=Disadvantage)
Customer’s perspective Manufacturer’s perspective A: Smaller volume of purchasing D: Lower sales volumes D: Increased risk of technological or
commercial obsoleteness
A: Lower demand for the company to be technically up-to-date and attentive to the voice of the customer D: Longer technology generations,
slowing down improvements of technology based on feedback from customers13
D: Longer technology generations, slowing down improvements of technology based on feedback from customers14
A: Potentially smaller cost per operational day.
A: More opportunities for revenues from services.
D: Extended warranty needed
A: High durability magnifies
reliability as a competitive advantages
I believe that most of these ideas are correct in a general sense. When applied to trucks though, it becomes a little more complex and requires further discussion and unfortunately a lot of information is needed to allow us to judge whether the above stated implications and other potential implications that I will present really apply. A simplified model of how I have come to view the issue of driveline durability is shown in the results section, while it is essentially conclusions based on internal interviews and general industry knowledge, although the ideas of Saleh (2006) are an integral part of the foundation.
13 Saleh mentions as an advantageous effect of shortening the design lifetime that there are more frequent iterations of getting products to the market, getting feedback from customers and solving issues before an improved product is put to the market. It seems reasonable to assume that this would translate to the opposite if the design lifetime is extended.
14 Same as above.
30
3 Methodology