Una oficina abierta con varios puestos de trabajo y pantallas
  1. Conglomerado
  2. Calaa Mind
  3. Consulting
  4. Lab and Robotics

01.2

What to measure, with which instrument, and with how much coverage.

The instrument is decided before the algorithm

Robotics builds the instrument, Lab writes the software that serves it, and Mind says what has to be measured and checks what was measured. The three are decided together, because an instrument that measures the wrong thing is not changed afterwards.

01.2.1

De dónde sale la medición propia

Robotics construye dos cosas. Un aparato que va dentro del vehículo y lee lo que el vehículo publica, y un satélite de observación que prepara la capa que ese aparato no puede ver. Lab escribe el software que recoge lo que los dos producen, entrena los modelos y mantiene la infraestructura que los sirve.

Lo que sale de ahí es la primera fuente del orden de preferencia de Mind. Es la única cuya metodología se conoce entera, porque la ha diseñado el grupo, y también la de menor cobertura, porque solo cubre lo que hay instrumentado.

What is instrumented

The most reliable source is the one that covers the least ground

The sector
RoboticsThe in-vehicle device
RoboticsThe satellite
What is not instrumented

Lab collects what both produce, trains the models and maintains the infrastructure that serves them.

The in-vehicle device
Reads what the vehicle publishes. Deep and narrow. It knows everything about what is instrumented and nothing about what is not.
The satellite
Prepares the layer the device cannot see. Wide and less fine, and it reaches where there is no device inside.
What is not instrumented
It is resolved with an external source, and it is the list Mind returns to Lab and Robotics to decide what is worth building next.

What falls outside is not left unanswered. It is resolved through the order of preference of sources, and every figure enters marked with the rung it came in by.

The field is the automotive sector and the two zones are what the group's instruments read. It ranks, it does not measure. It says which part is instrumented and which is not, not how much each one covers.

01.2.2

Qué le pide Mind a cada una

Tres peticiones concretas, y las tres se hacen en el diseño y no después.

  1. 01

    A Robotics

    El instrumento

    Qué campo hay que leer y con qué frecuencia, para que la serie sirva para comparar y no solo para observar. Una medición tomada cada semana y otra tomada cada trimestre no se pueden cruzar, y eso se decide al fabricar, no al analizar.

  2. 02

    A Lab

    El recorrido del dato

    Que la fuente, la fecha y el método viajen con el dato en cada salto. Un dato que pierde su origen al agregarse deja de ser comprobable, y entonces vale lo mismo que uno comprado.

  3. 03

    Lo que Mind devuelve

    Dónde falta cobertura

    Qué parte del sector no está instrumentada y hay que resolver con fuente externa. Es la lista que decide qué merece la pena construir después, y la única que sale de haber intentado medirlo.

The two cadences

Two series with different steps almost never coincide

One measurement every week

Short step

Where only one does

Another every quarter

Long step

Decided when buildingNot when analysing

The frequency is decided when the instrument is built, not when its output is analysed. A series taken at the wrong step is not corrected afterwards, it is taken again.

The same time read by two instruments at different frequencies. The links mark the moments when both have data at once. It ranks the calendar; it does not say how much information each mark holds.

Tres personas revisan un análisis en un portátil, en una sala con vistas a la ciudad

01.2In the design

Una medición tomada cada semana y otra cada trimestre no se pueden cruzar. Eso se decide al fabricar.

The question of what to measure comes before the instrument, not after watching it run.

01.2.3

Qué recibe un equipo de ingeniería

Lo mismo que cualquier otro trabajo de Mind, con una diferencia de formato que se pide siempre y casi nadie entrega.

  1. The answer, first

    Whoever reads only the first page has to be able to decide.

    The question answered on one page, with the recommendation and the confidence level. The rest of the report exists so that page can be checked if they choose to.

  2. The reasoning, with the discards

    A report that only shows the path that worked cannot be argued with.

    How the recommendation is reached from the data, with the alternatives left out and the reason for each discard. That is where an engineering team finds whether its case was considered.

  3. A separate data sheet

    A figure that cannot be copied with its backing gets copied without it.

    The figures quotable separately, each with its source behind it. It is what is always asked for and almost never delivered, and without it the backing is lost at the first jump to another deck.

01.2.4

Por qué se decide junto

Un satélite tiene fecha de lanzamiento y un aparato tiene serie de fabricación. Lo que no se pidió en el diseño no se añade en la siguiente versión del software, y entonces el sector se mide con lo que se puede en vez de con lo que hacía falta. Por eso la pregunta de qué medir llega antes que el instrumento.

How long it can still be changed

The three pieces close at different moments

  1. The software

    Fixed at

    Each release

    Change afterwards · Yes

  2. The in-vehicle device

    Fixed at

    The manufacturing run

    Change afterwards · No

  3. The satellite

    Fixed at

    The launch date

    Change afterwards · No

The question of what to measure has to arrive before the first door closes. After that, the sector is measured with what can be measured instead of what was needed.

When each piece of the instrument is fixed and whether it admits a change afterwards. It compares decision moments, not timescales or cost.