Lesson 2 of 16 · Automotive SPICE, the vital 20 %
Reading a process
Every one of the 32 processes is written to the same template. Learn to read one and you can read them all, and you can judge whether a team reaches level 1 on it.
PA 1.1 L · PA 2.1 L · PA 2.2 F. Which level?
"Every Jira story must have an acceptance-criteria field filled in." WHAT, HOW or DOING?
The anatomy of a process
Chapter 4 describes every process with the same four parts[Table 10, p. 28]:
| Part | What it is | How to use it |
|---|---|---|
| Purpose | One sentence: why the process exists. | The test of last resort: is this achieved? |
| Outcomes | A numbered list of observable results. | This is what PA 1.1 rates. |
| Base practices | BP1…BPn: activities that achieve the outcomes. Process-specific. | Activity-oriented evidence: "do they do this?" |
| Output information items | Numbered items such as 08-56 Schedule, described in Annex B. | Result-oriented evidence: "what should I find?" |
A table at the end of each process says which BP and which information item support which outcome. That table is the assessor's map.
PA 1.1 in one line
Level 1 has a single attribute with a single generic practice: GP 1.1.1 "Achieve the process outcomes", which means achieving the intent of the base practices and producing work products that evidence the outcomes[§5.2.1, p. 100]. So rating PA 1.1 comes down to judging whether the outcomes are achieved, using the base practices and information items as indicators. They are not a mandatory checklist[p. 23].
Information items are not documents to write
This is the most common beginner mistake. An information item is what the assessor looks for. A work product is what the team actually produced. There is no 1:1 relationship: one Jira board can contain the characteristics of several information items, and one information item can be spread over many artefacts[§3.3.2, pp. 24–25]. The characteristics in Annex B "shall not be interpreted as a required structure"[p. 24]. Evidence comes from two sources only: work products (including repository content) and testimony from the people doing the work[p. 24].
A ticket in Jira is a work product. The Guidelines say so explicitly: change requests or problem reports held in a workflow tool count[Guidelines §4.2.1, p. 246]. Nobody needs a Word document called "Change Request".
A worked example: MAN.3 Project Management
Purpose: "to identify and control the activities, and to establish resources necessary for a project to develop a product, in the context of the project's requirements and constraints"[p. 89].
| Outcome | Base practices the PAM maps to it |
|---|---|
| 1 · Scope of work defined | BP1 Scope of work · BP2 Lifecycle |
| 2 · Feasibility evaluated | BP2 Lifecycle · BP3 Feasibility · BP5 Estimates and resources |
| 3 · Activities and resources sized and estimated | BP4 Work packages · BP5 · BP6 Skills · BP7 Interfaces · BP9 Consistency |
| 4 · Interfaces identified and monitored | BP4 · BP7 · BP9 |
| 5 · Plans developed, implemented, maintained | BP4 · BP9 |
| 6 · Progress monitored and reported | BP8 Schedule · BP10 Review and report progress |
| 7 · Adjustment when goals are not achieved | BP9 |
From the mapping table on pp. 90–91; most BPs support several outcomes. Output information items include 08-53 Scope of work, 14-10 Work package, 08-56 Schedule and 15-06 Project status.
What the assessor found at Seat-Heat
The Guidelines contain rating rules written for exactly this situation. One is [AGE.RL.2]: "If remaining effort for features, which are to be delivered in the current or next release, is not estimated, then MAN.3.BP5 shall be downrated"[Guidelines §2.5.2.1, p. 77]. Lesson 8 covers rating rules in full.
Practice
A. Which base practice does this evidence speak to?
The project charter with goals and the variants in and out of scope.
The CAN-signal dependency on the HVAC team.
Sprint reviews every two weeks and a monthly status mail to the OEM.
No estimate of the effort still needed for the next two releases.
B. Traps
Must Seat-Heat produce a document titled "08-56 Schedule" to satisfy BP8?
An engineer explains in the interview how feasibility was checked. What kind of evidence is that?
Across the whole dossier above, what would you rate MAN.3 PA 1.1?
C. From memory
Name the four parts of a process description, and say which one PA 1.1 rates.
Purpose, outcomes, base practices and output information items. PA 1.1 rates whether the outcomes are achieved, judged through the base practices (activity indicators) and information items (result indicators).
Your win: you can open any process, map evidence to its base practices and give a defensible PA 1.1 rating. Next: which processes an assessment actually looks at, so you know which dozen of the 32 to learn.
To keep it: import this lesson's Anki deck (how).
Read next
MAN.3 in the PAM, pp. 89–90. Two pages, and the process nearly every assessment includes. Then look up 08-56 Schedule and 15-06 Project status in Annex B to see what "characteristics" look like.
Unsure why a piece of evidence maps to one BP rather than another? Ask your teacher in the chat.