← Every exhibit

Figure 1From “What is AI demand forecasting? A forecast is a system, not an oracle”

The AI forecast control map

Automation can produce a number; a governed forecast also declares the target, cutoff, version, override, owner, and later evaluation.

Cite Embed

Figure 1The AI forecast control map

A schematic mirror chart compares an automation-only path with a bounded human-AI forecast path. Both declare target, data cutoff, model version, and baseline. The bounded path additionally records override boundary, reason, owner, and later actual. Values are coded field presence, not performance.← Declared fieldsProcess path →DefinitionInformationModelForecastGovernanceEvaluation1 coded presence1 coded presenceTarget and unit1 coded presence1 coded presenceData cutoff1 coded presence1 coded presenceModel version1 coded presence1 coded presenceBaseline value0 coded presence1 coded presenceOverride boundary0 coded presence1 coded presenceReason and owner0 coded presence1 coded presenceLater actual

Source: Author's schematic comparison grounded in Shrestha, Ben-Menahem and von Krogh (2019) and Raisch and Krakowski (2021). Values are coded control-field presence, not model performance, accuracy, or commercial results.

PNG4:3, 1600 × 1200, with the title, the source line and the attribution set in the image isoglu-2026-what-is-ai-demand-forecasting-fig05.png

Source and evidence

Each line names a claim related to this exhibit, its source, and how it was checked.

  • A SBK19-C1 The five, named: "five key contingency factors specificity of the decision search space interpretability of the decisionmaking process and outcome size of the alternative set decision making speed and replicability"Shrestha, Ben-Menahem & von Krogh. (2019)
  • A SBK19-C2 The factors are a comparison frame, not a ranking: the article "identifies the idiosyncrasies of human and ai based decision making along five key contingency factors", so different combinations imply different splits of the workShrestha, Ben-Menahem & von Krogh. (2019)
  • A RAI21-C1 The two concepts, defined: "whereas automation implies that machines take over a human task, augmentation means that humans collaborate closely with machines to perform a task"Raisch & Krakowski. (2021)
  • A RAI21-C3 "if organizations adopt a broader perspective comprising both automation and augmentation they could deal with the tension and achieve comp"lementarities. A conceptual paper in AMR, so this is argued, not measuredRaisch & Krakowski. (2021)
  • B V11-OWN-C1 AI demand forecasting is a forecast process in which a model uses declared data and horizon to produce a demand estimate for a named unitAuthor framework grounded in SBK19-C1 and RAI21-C1
  • B V11-OWN-C2 A demand-forecasting record should preserve baseline model, input snapshot, horizon, output, confidence or interval, override, reason, and later actualAuthor operating framework
  • B V11-OWN-C3 Automation, augmentation, human judgment, forecast output, and realized demand are separate objectsAuthor taxonomy grounded in RAI21-C3
  • B V11-OWN-C4 AI forecast quality must be compared against a declared baseline, loss function, segment, maturity window, and data cutoffAuthor evaluation framework

Grades: A, verified against the printed page of the primary source · B, primary source, text layer only · C, authoritative secondary · D, reported.

From the essay

Cite or embed

APA 7

Isoglu, S. (2026). The AI forecast control map [Figure]. isoglu.com. https://isoglu.com/insights/exhibits/isoglu-2026-what-is-ai-demand-forecasting-fig05/

BibTeX
@misc{isoglu-2026-what-is-ai-demand-forecasting-fig05,
  author = {Isoglu, Sinan},
  title = {{The AI forecast control map}},
  year = {2026},
  howpublished = {Figure 1, isoglu.com},
  url = {https://isoglu.com/insights/exhibits/isoglu-2026-what-is-ai-demand-forecasting-fig05/},
  note = {From the essay {What is AI demand forecasting? A forecast is a system, not an oracle}}
}
Embed this exhibit

Copy this HTML to use the PNG exhibit on your site with its source line and a link back to the exhibit page.

<figure>
  <a href="https://isoglu.com/insights/exhibits/isoglu-2026-what-is-ai-demand-forecasting-fig05/">
    <img src="https://isoglu.com/social/isoglu-2026-what-is-ai-demand-forecasting-fig05.png" alt="Figure 1: The AI forecast control map" width="3200" height="2400">
  </a>
  <figcaption>Figure 1: The AI forecast control map. Source: Author&#39;s schematic comparison grounded in Shrestha, Ben-Menahem and von Krogh (2019) and Raisch and Krakowski (2021). Values are coded control-field presence, not model performance, accuracy, or commercial results. <a href="https://isoglu.com/insights/exhibits/isoglu-2026-what-is-ai-demand-forecasting-fig05/">View exhibit</a>.</figcaption>
</figure>

Related exhibits

Figure 1Figure

The human-machine division-of-labor card

Human and machine should divide the work

Table 4Table

What evidence should a sales forecast carry?

What is AI demand forecasting? A forecast is a system, not an oracle

Table 1Table

What does AI demand forecasting mean?

What is AI demand forecasting? A forecast is a system, not an oracle

Every exhibit

Contact

Bring the question you're working on.

Research, a guest lecture, or a commercial question.