Short pieces on the problems we keep meeting, and what we have found underneath them. Written for the people who have to sign, not for a conference. All of these are in draft — ask and we will send you one before it is published.
Reading data is easy and nobody is afraid of it. Writing is where every pilot stops, because the first question is not how accurate the model is — it is what it did, who allowed it, and whether it can be undone. Most architectures cannot answer, because the record of what happened is written separately from the thing that happened. Two sources of truth that drift.
ASK FOR THIS NOTEA stay crossing three wards is three ward-stays. Credit it whole to the ward that admitted, as almost every spreadsheet does, and you systematically overstate wards that admit and understate wards that receive — then fund beds against those numbers.
ASK FOR ITSeven questions, none of them about model accuracy. If a vendor cannot answer all seven in a single meeting, the pilot will stall at exactly the point where it starts being worth money.
ASK FOR ITA hospital possesses patients. A mine possesses production equipment. A logistics network possesses shipments. What one unchanged engine running a hospital ward and a wastewater treatment works taught us about which parts of a business are actually industry-specific.
ASK FOR ITThese are written as the work happens rather than to a publishing calendar. Tell us which one is useful to you and we will send it when it is ready — or send you the draft now, if it would settle an argument you are currently having.