Ask a supplier what a machine costs and you will get a number within a day. Ask what a part costs on that machine, in your shop, with your labour and your volumes, and the answer usually arrives as a slide showing "low running cost".
That gap is where most bad laser investments are made. The purchase decision is nearly always framed as a machine price comparison, when the thing that actually determines whether the investment works is the cost per part at your real utilisation — and that number depends far more on your operation than on the machine's specification.
This article does not give prices. Any figure printed here would be wrong for most readers, because wages, power, floor space and material costs vary enormously between markets. What it gives is the structure: which buckets to fill, where to get each figure honestly, and which parts of the arithmetic people habitually get wrong.
Costus per partus est summa septem rerum. Plurimae quotatiores includunt duas aut tres earum.
Tres horum saepe praetermittuntur, et quaelibet eorum materialiter mutat responsionem.
Spatium prope semper praetermittitur, quia videtur esse pretium iam perditum. Non est: machina quae spatium in officina urbana occupat aliquid consumit cuius pretium in mercatu est, et secunda machina semper plus spatii consumit quam prima, quia depositio materiae et reliquiarum cum ea crescit.
Consumptibilia solent tantum ex maximo articulo aestimari — tubo — et optica, effusoria et filtrorum mentionem neglegere. Filtra praesertim sunt repetitus impensus qui cum quantitate materiae tractata crescit, non cum horis machinae; ergo officinae quae spumam aut tabulam obductam recidunt, illos multo saepius mutant quam expectant.
Cede est le seau qui distingue une boutique comprenant bien son activité d’une autre qui ne la comprend pas. Le matériau constitue généralement le coût unique le plus élevé dans le travail au laser, et l’écart entre un bon agencement (nesting) et un agencement négligé pour le même travail représente un pourcentage réel du coût de la pièce. Le guide sur la largeur de coupe (kerf) et l’agencement traite de la géométrie ; cet article insiste uniquement sur le fait que cette valeur doit être mesurée, non supposée.
La formule standard est : coût par pièce = taux horaire-machine × heures par pièce + coût du matériau par pièce.
C’est une bonne formule pour une usine produisant un seul produit pendant des mois. Elle est trompeuse dans les travaux que la plupart des boutiques laser effectuent réellement, pour une raison : la préparation n’est pas proportionnelle au temps de découpe.
Considérons deux travaux sur la même machine :
The machine-hour rate attributes the same cost to both, because it only counts cutting hours. But Job B consumed ten times the loading, nesting, alignment, test-cutting and inspection. If the shop prices both with the same rate, Job B is being sold at a loss — and it is usually the job that felt like more work, which is the shop's instinct telling the truth while the spreadsheet disagrees.
The practical fix is to cost by oRDER , then divide:
Hoc est mutatio unicuique utilissima quam plerique officinae possunt adhibere in pretiis suis: addere praebitum praeparationis pro singula ordine linea , quod aestimatur ex vero tempore necessario, deinde autem pretium machinae tantum tempus operationis amplectitur. Ita resulat ut parvi et varii ordines honeste pretientur, non subsidiantur per longas series — quod etiam est causa cur multae officinae simul sentiantur occupatae et inprosperae.
Haec supra cistae sunt fixae et variabiles, atque fixae super horas quotquot machina vere operatur distribuuntur. Ita utilisatio potentissimum numerum in tota calculatione efficit — et eum qui saepissime supponitur potius quam metitur.
Arithmetica simplex est et aperte enuntianda: pretium fixum pro hora = pretium fixum annuum ÷ horae machinae annuae. Reduca horas ad un tertium et costum fixum per horam fere triplicatur. Nihil de machina mutatum est; tantum quantitas operis quae per eam transit.
Duae sincerae monitiones de sectando utilisationem.
Capacitas inactiva non est capacitas libera. Machina quae inactiva sedet adhuc spatium, depreciationem et assurantiam consumit. Officina quae capacitatem emerit "pro incremento" et eam in quinta parte horarum suarum exerceat, totum pretium fixum solvit pro quinta parte producti — quod est praecise modus quo machina quae videri poterat esse utilis fit causa cur negotium nihil aliud investire possit.
Secundum turnum esse solet minime pretiosam capacitatem disponibilem , quia pretium fixum iam solutum est. Antequam secundam machinam emas, sincera ratio est an prima machina ulterius onerari possit — et saepe vinculum non est machina sed operator, mensa finienda, aut expectationes clientis de tempore ducendi.
Reddito di ritorno è solitamente presentato come "prezzo della macchina ÷ risparmio mensile". Quella versione nasconde tre cose e favorisce quasi ogni proposta.
Un calcolo difendibile del reddito di ritorno ha questa forma:
Costo annuale dell'investimento = ammortamento + interesse o costo di finanziamento + spazio + energia + consumabili + manutenzione + assicurazione + lavoro aggiuntivo necessario per farla funzionare.
Beneficio annuale = quanto non paghi più a qualcun altro, più quanto puoi ora vendere ma prima non potevi, meno i materiali e il lavoro che ora consumi per farlo internamente.
Periodus Reparationis = costo annuale dell'investimento ÷ beneficio netto annuale.
Tre correzioni che modificano il risultato:
Confronta con l'alternativa reale, non con zero. Se il lavoro è attualmente esternalizzato, il beneficio è la fattura esterna meno il tuo costo interno — non l'intera fattura. Se il lavoro non viene attualmente svolto affatto, il beneficio è speculativo e va etichettato come tale.
Include the labour you are adding. A machine that replaces an outsourced invoice but requires a full-time operator may be neutral, and a proposal that omits the operator is not a calculation, it is a hope.
Use the volume you can actually sell, not the machine's capacity. Capacity-based payback is the most common error in equipment proposals: it assumes the machine runs at its ceiling from month one, which almost no shop achieves.
A list worth taking to any negotiation, because each of these is real and each is usually discovered after installation.
The learning curve deserves its own line. A new machine in an existing shop does not reach its expected output on the first day; the first weeks produce scrap, slower nesting and more questions. A payback model that starts at full productivity is wrong by exactly that ramp — and the ramp is longer for shops new to the process than for shops adding capacity.
The point of this article is not to hand over a number but to make the calculation unavoidable. The table below is the minimum honest structure; fill the right-hand column from your own records, not from a supplier's presentation.
Si una proposta non potest in hoc structura cum tuis numeris impleri, adhuc non est casus commercialis — est tantum quotatio.
Ubi arithmetica exstat, quattuor decisiones responsum habent, et unaquaeque habet diversum responsum secundum numeros.
Externe conducere an emere? Sub certo volumine, conductio externa vincit, quia costum fixum omnino vitatur. Super eo, proprium facere vincit, quia costum fixum tenuius diffunditur. Punctum intersectionis pendet a tuis voluminibus et a pretio locali conductoris externi — quare nullum articulum id tibi dare potest.
Nunc emere an maius postea emere? Machina minor cum alta utilisatione saepius vincit machinam maiorem quae in fractione suorum horarum operatur. Exceptio est opus quod physice non in ea capitur, ubi decisio est de facultate, non de pretio.
Unum momentum an duo? Secunda turnus est communiter optima deciso pro capacitate, quia pretium fixum iam solutum est. Limitatio saepe est capacitas finiendi aut disponibilitas operatorum, non machina.
Addere secundam machinam an secundum processum? Si locus angustus est in sectione, secunda sectrix iuvat. Si locus angustus est in finitione, in ligatura marginum, aut in conpositione, secunda sectrix tantum maiorem catenam ante eundem locum angustum aedificat. Computatio totius lineae, non machinae, hoc errorem prohibet.
A laser is not cheap or expensive in the abstract; it is cheap or expensive per part, at your utilisation, in your material mix . The only way to know which is to build the arithmetic yourself — seven buckets, an honest hour count, and a setup charge that reflects what small orders actually consume.
Suppliers who resist that conversation are telling you something useful. A supplier who will sit down with your numbers, including the ones that make their machine look less attractive, is the one worth buying from — because the machine will still be in your shop long after the quotation has been filed.
The calculation is not complicated. It is just uncomfortable, because it forces the two questions most purchase decisions avoid: how much will this machine actually run, and what will I stop paying someone else to do?
Nuntii Calidi2026-11-16
2026-11-12
2026-11-10
2026-11-09
2026-11-06
2026-11-05