In the previous article (see Reliability in PBCs – Part 1) we defined how reliability of a system or service could be used in a Performance Based Contract (PBC) including defining whether we were measuring success or failure, and whether the failure resulted in a mission [critical] failure or not (defined as a logistics failure). In this article we are going to look at the detail needed to measure reliability from a contractual perspective.
Previously we defined those elements that would result in either a mission or logistics failure. However, from a contractual perspective and linked to performance management, who is attributed to (owns) this failure, Buyer or Seller? Typically, this refers to either a chargeable (to the seller) or non-chargeable (to the buyer) failure. Unfortunately, determining whether a failure is chargeable or non-chargeable is not simple, especially when dealing with complex systems and commercial arrangements (e.g. multi-party contracts). As illustration of this consider the following scenario.
The buyer receives serviceable (working) vehicles from the seller that are then driven by the buyer’s employees to deliver buyer’s goods. The seller is responsible for ensuring that there are enough vehicles available to the buyer with all maintenance completed and the vehicles are “roadworthy” (i.e. no safety or mechanical issues such as working lights).
Now consider the buyer has failed to make a delivery due to the vehicle not working. Is this a chargeable failure? If the failure was due to a mechanical issue (e.g. engine not starting) then this is simple; chargeable failure to the seller. However, what if the vehicle was in an accident that was the fault of the buyer’s employee? Or the buyer’s driver did not put fuel in the vehicle? What then? What if a third-party caused the accident (e.g. neither the buyer nor seller)? Alternatively, what if the vehicle was operating outside it’s normal role (e.g. carrying twice the maximum design weight making the brakes less effective) or in a different environment (e.g. extreme cold resulting in an icy road). What now? Is the failure chargeable or non-chargeable?
In a previous article (see Setting the Performance Levels (Part 3)) I discussed criticality of Configuration, Role and Environment (CRE) in setting performance levels. In failure sentencing (the name given in the Reliability Engineering community for this process), we should also consider the CRE. While it is possible to transfer all risks from buyer to seller, in many circumstances this risk transfer is unaffordable for the buyer.
The alternative is to specify all these exclusions in the PBC. Indeed many years ago my colleagues and I would try, sometime resulting in up to 3 pages of possible exclusions like the ones above. However, explicitly considering and documenting is neither effective or efficient since it is impossible to capture all possible event. Additionally, the drafting and management cost to both buyer and seller becomes too high.
To avoid this in recent years my colleagues and I have started either:
- defined general sentencing principles (as opposed to specific rules); or
- used “all in” performance measures that are not adjudicated (e.g. no failure sentencing is undertaken). See When a KPI is not a KPI for further details.
The key to which of these 2 types you use is highly dependent on whether the commercial relationship can include “all in” performance measures as part of the performance measures hierarchy. In my experience many contracts are more “master-servant” in nature (i.e. the buyer will tell the seller what to do and how to do it) as opposed to a more partnering style that emphasises a “shared destiny” approach. However, where the commercial relationship is more partnering and collaborative in nature, the “all in” performance measures, especially when linked to incentive (see PBC Incentive Regime Part 1 and Part 2), are very powerful drivers of positive outcomes for both buyer and seller.
In summary, like availability performance measures, reliability performance measures should be included in many PBC performance measure hierarchies as they represent the goal of many PBC buyers and sellers. However, care must be taken when designing these reliability performance measures to make sure that they accurately reflect the need and take into account the specific CRE of the commercial arrangement. Moreover, unless the commercial relationship reflects a more partnering approach, the use of “all in” performance measures will be difficult to include.