CB2500 Chap.4 Process Quality, Repositories and Automation
Process Quality, Repositories and Automation
Three claims in one sentence
The week two materials compress the process-quality argument into a single line: keeping the same fact in two stores tends to make a process inefficient, whereas holding it once costs less, produces fewer mistakes and serves both teams equally well. Three distinct claims are being made, and only the third is surprising. Lower cost is arithmetic once the reconciliation effort is counted.
Fewer errors follows because copies that are kept in step by hand drift between reconciliations. The claim that carries the argument is that nothing is lost, and it answers the objection every affected team raises, which is that they will lose the view they need.
A view is not a copy: two functions can read one record through different presentations without either owning a separate version of the truth.
Efficiency and effectiveness answer different questions
Keep the two words apart. Efficiency asks whether the same result is produced with less input. Effectiveness asks whether the result is the right one.
A process can become dramatically more efficient and less effective at the same time, which is what happens when a control is removed to save a step.
When a question asks how an information system improves process quality, say which of the two is moving and whether the other is being held constant.
Automation is a boundary, and the boundary is a decision
The lecture closes the topic with the bridge back to the five components. Actors and instructions sit on the human side, hardware and software on the computer side, and data and procedures join them.
Automation is then defined as work moving from the human side to the computer side, and the analyst is described as settling how far automation should go and what the move costs in equipment, programs, networking, training and staffing.
The decision is not whether to automate but where to put the line, and what moving it costs across all five components rather than in software alone.
Three questions that place the line
How variable are the cases, because highly repetitive work automates well while every-case-different work pushes cost into rules for exceptions.
What does an error cost, because where a wrong outcome is expensive and hard to reverse a person in the path is a control rather than a delay. And what is the investment in full, since the licence is rarely the largest line and procedure rewriting and training usually are.
What this chapter covers
- 01
Two repositories against one, as three separate claims
- 02
Why a shared record is not a loss of view
- 03
Reconciliation as the hidden recurring cost
- 04
Efficiency against effectiveness
- 05
The human side and computer side bridge
- 06
Automation as work crossing the boundary
- 07
Case variability, error cost and full investment
- 08
The exception that stays with a person
Consolidate a store and state what consolidation cannot fix
- 3Trace the symptom to the arrangement rather than to a person.
- 4Give the change and state all three consequences separately.
- 3Name what remains unfixed and say which chapter it belongs to.
Key terms
- Process Quality
- How well a process performs, judged separately on the resources it consumes and on whether its outcome is the right one.
- Efficiency
- Producing the same outcome with less input, measured in steps, time, rework or cost.
- Effectiveness
- Producing the outcome the organisation actually needs, independent of how much it costs to produce.
- Reconciliation
- Periodic work to bring two copies of the same fact back into agreement, and a recurring cost of holding two stores.
- Automation Boundary
- The chosen dividing line between work performed by people and work performed by machines within one process.
- Integrated Repository
- A single store read and written by every activity needing the fact it holds, replacing per-function copies.
Process Quality, Repositories and Automation FAQ
Why is consolidating two stores described as no loss of effectiveness?
Because each team keeps the presentation it needs while giving up the private copy underneath it. The objection people raise is that a shared store forces everyone to look at the same screen, which is not what is being proposed. In the improved version of the lecture example the shared store also carries performance data, so the consolidation adds a capability rather than removing one.
How do I decide how far to automate a process?
Ask three questions and answer all of them. How variable are the cases, because rules for exceptions are where the cost hides. What does a wrong outcome cost and how reversible is it, because an expensive irreversible outcome is a reason to keep a person in the path. And what is the total investment across hardware, software, networking, training and manpower, which is the list the lecture itself gives.
Is more efficiency always an improvement in this course?
Not on its own. Efficiency and effectiveness are separate, and a change that removes a step may remove a control with it. An answer that reports a time saving without saying whether the outcome is still the right one has answered half the question, and a marker reading for process quality will notice which half is missing.
Exam move
Take one process you can observe and count its stores: every place the same fact is written. Then write what reconciles them and how often, because that effort is the cost nobody invoices. For the automation half, pick a single activity and write the three placing questions with an answer each, then write the exception that would still need a person.
A recommendation without that exception reads as untested, and practising it now makes it automatic under time pressure.
Working through Process Quality, Repositories and Automation in CB2500? Sia is AskSia’s AI Information Technology tutor — ask any CB2500 Process Quality, Repositories and Automation question and get a clear, step-by-step explanation grounded in how CB2500 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
