600+
Projects Designed
3M+
m² of Slabs
7
Countries

What is OptiSlab?

OptiSlab is a proprietary software developed by LTPisos for the structural design of concrete slabs-on-ground in industrial facilities. Its objective is to predict the actual stresses experienced by the concrete under its loads and specific conditions, and compare them with the true allowable capacity of the slab.

Unlike conventional methods—which use closed-form equations for idealized infinite plates—OptiSlab employs Artificial Neural Networks (ANN) trained with nearly one million simulations using the Finite Element Model Islab2000. The result is FEM-level accuracy analysis in minimal time.

It is important to note that the beam test for determining MOR (Modulus of Rupture) is insufficient to correctly predict the actual capacity of the slab. This method tends to deliver an underestimated result, as it does not consider the specific conditions of the slab in the field or the actual effects of loading. Therefore, OptiSlab corrects this value by calibrating it with empirical testing, to obtain an accurate response about slab capacity and ensure structural safety and efficiency in industrial applications.

OptiSlab does not seek merely to be an alternative to what is proposed by ACI 360 or TR34. It surpasses them by correcting the limitations that their own authors acknowledge: the inability to model actual slab curling, the effect of panel size, and the use of MOR as a proxy for actual floor capacity.

Developed by the company and continuously improved since 2012, OptiSlab is the technical foundation that defines LTPisos' market differentiation.

What no conventional method models

OptiSlab is the only method that simultaneously considers:

  • Actual slab curling (equivalent temperature differential incorporated)
  • Effect of actual panel size (1.4 m to 40 m)
  • Maximum stress on upper and lower fibers
  • True slab capacity, not laboratory MOR
  • Load transfer efficiency factor (LTE) at joints
  • Effective k-value multilayer by Palmer-Barber method
  • Punching Shear Analysis with fiber contribution
  • Safety factor 0.80 static / 0.45 dynamic (forklift fatigue)
  • Comparison of alternative floor types

Why conventional methods are not sufficient

ACI 360 and TR34 are widely used standards, but their own authors warn of fundamental limitations that OptiSlab overcomes.

ACI 360 is based on Westergaard's elastic solution for a semi-infinite plate on Winkler foundation. TR34 is based on Meyerhof's plastic theory for cracked sections. Neither considers the actual panel size nor curling—the two variables that most influence stresses where nearly all industrial floor failures initiate: at the joints.

Design Criterion ACI 360
(Westergaard)
TR34
(Meyerhof)
OptiSlab
(FEM + ANN)
Models slab curling ✗ No ✗ No ✓ Yes — Tbi incorporated
Considers actual panel size ✗ No — semi-infinite plate ✗ No — semi-infinite plate ✓ Yes — 1.4 m to 40 m
Maximum stress above and below slab at critical positions Partial: Edge, center or corner ✗ No ✓ Yes — both simultaneously
True slab capacity (not MOR) ✗ Uses MOR directly ✗ Post-crack only ✓ Yes — corrected by C1, C2 factors
Load transfer efficiency (LTE) ✗ Not integrated ✗ Not integrated ✓ Yes — aggregate interlock and dowel model
Multilayer effective k-value of soil ✗ Simple k ✗ Simple k ✓ Yes — Palmer-Barber method
Punching Shear Analysis with fibers and edges Partial Partial ✓ Yes
What ACI 360 says about its own methods: The standard explicitly warns that the three design methods (PCA, WRI, COE) "give incorrect slab thickness results when the slab is not in contact with the subgrade." As k increases, the methods allow thinner slabs—but higher k produces longer unsupported edges and reduces actual capacity at the joints. OptiSlab is, in part, a direct response to that warning.

The OptiSlab engine: Islab2000 + Neural Networks

A mechanistic approach that transforms millions of FEM simulations into an instant, precise design tool

Step 1
🔬

Islab2000 Simulations

Nearly one million FEM analyses with systematic variation of thickness, panel size, k, LTE, loads, fibers, and incorporated differential temperature. Each run generates the elastic response of concrete under specific, real-world conditions.

Step 2
🧠

ANN Training

Artificial Neural Networks learn to reproduce FEM's elastic response with error less than 5%. What FEM takes hours to calculate, OptiSlab delivers in minimal time—without sacrificing accuracy.

Step 3
📐

Real project design

The engineer enters the actual project conditions. OptiSlab returns stresses, corrected allowable capacity, and safety factor for each structural alternative considered.

OptiSlab Artificial Neural Networks (ANN)

ANN for Racks

Trained on a massive set of Islab2000 simulations for static rack configurations and counterweight forklifts, considering load position relative to joint, LTE, and panel geometry.

ANN for Forklifts and Cranes

Developed by Rodrigo Molina M. (Universidad de Los Andes, 2021). Extends OptiSlab to the full range of material handling equipment: reach trucks, trilateral trucks, order pickers, boom hoists, and generic loads.

Point Loads

For situations with specialized machinery, temporary storage of heavy equipment, or maintenance vehicles. The user enters the load magnitude (kg), contact area dimensions (cm), and position (center or edge of panel). Location significantly influences the result: placing it at the edge generates higher stresses and a more conservative analysis.

Distributed Loads

Mode 1: Uniform load across the entire panel—represents areas where products completely cover the space. Critical stress appears on the lower face, near the panel center.

Mode 2: Two loaded zones with central aisle for forklifts (width 1–2.5 m). Produces more complex stress distribution: loaded zones push downward and the unloaded aisle creates an inverted bending effect.

Curling without load

Analysis of the effect of curling on the slab without applied load. Considers incorporated differential equivalent temperature (Tbi) to model actual slab behavior in the field.

Punching Shear Analysis

In addition to verifying bending, OptiSlab performs punching control for all rack loading cases. Capacity is expressed in kilograms (maximum load before failure). For acceptable design, both bending and punching safety factors must exceed 1.0. In many practical cases, punching determines the minimum required thickness for high-load racks.

Three-module architecture

OptiSlab organizes design in a logical sequence: first loads, then structure, and finally automated calculations

1

Module 1 — Load Description

Complete library of industrial equipment and loads. Users select the actual project equipment.

  • Racks: load per leg, height, leg spacing
  • Counterweight forklifts (preloaded models)
  • Reach trucks, trilateral trucks, order pickers
  • Boom hoists and electric pallet jacks
  • Generic point and distributed loads
  • Punching Shear Analysis under rack base
2

Module 2 — Structural Description

Defines the actual conditions of the floor and foundation soil.

  • Slab thickness and typology (ER, RC, PT, fiber)
  • Panel size and joint layout
  • Incorporated differential temperature (Tbi)
  • Concrete properties and fiber strength (ASTM C1609)
  • Multilayer support system (CBR / E per layer)
  • Joint type and load transfer efficiency (LTE)
3

Module 3 — Intermediate Calculations

OptiSlab generates all design results without additional manual intervention.

  • Multilayer effective k-value modulus (Palmer-Barber)
  • LTE efficiency for each joint type
  • True floor capacity—allowable stress is calculated with calibrated factors C1 (thickness) and C2 (ASTM C1609 fiber), without using MOR directly
  • Safety factor for static and dynamic loading
  • Verification of Punching Shear with fibers

OptiSlab History

Over a decade of continuous development

2010

Multilayer module and LTE

Implementation of Palmer-Barber method for multilayer k calculation. Load transfer efficiency model at joints.

2012

Version 1.0 — The first prototype (OptiFloor)

First ANNs trained with Islab2000 simulations for rack loads. Start of Optimized Slab methodology.

2014–18

Expansion to mobile equipment

Models for counterweight forklifts and narrow-aisle equipment.

2021

OptiLoad — ANN for equipment (Universidad de Los Andes)

Rodrigo Molina M. develops neural networks for the full range of material handling equipment.

2022–24

Version 4.x — ~1 million simulations

ANNs trained with nearly one million Islab2000 runs. Punching with fibers. Presentation at Eurasphalt & Eurobitume 2024.

2026

SLABS — The platform of the future

Web platform integrating OptiSlab with project management tools, multi-user access, and continuous learning.

More than a decade of results

Continuous development from 2010 to today

From the first prototype in 2012 to the current version with nearly one million FEM simulations, OptiSlab has continuously evolved to reflect real-world industrial floor conditions in the field.

  • Over 600 projects designed across 7 countries
  • Over 5 million m² of slabs calculated
  • Error less than 5% versus direct FEM (Islab2000)
  • Presented at Eurasphalt & Eurobitume 2024
  • Web platform SLABS in development (2026)

Want to see OptiSlab in action?

Let's talk about your project and show you how OptiSlab improves design decisions for your industrial floor.

Request Technical Consultation View Projects