IHPC Tech Hub

IHPC Tech Hub showcases IHPC's suite of in-house developed applications, tools or technology to help you unlock the possibilities to overcome business challenges. Through providing valuable insights, you can now predict and shape the commercial outcomes, automate processes, and free up resources for repetitive and labour-intensive tasks. 

Discover the power of computational modelling, simulation and AI that brings about positive impact to your business. 

Integrated Smart Farm Model (iSAM)

Today, more than 50% of the world's population lives in cities. Projection from the United Nations#1 shows that two-thirds of the population on earth in 2050 will be living in urbanised metropolitan areas. Hence traditional agriculture could no longer support the need for fresh and healthy food for megacities. One potential solution is Urban Farming: growing crops in big cities. 

The advantages offered by these urban farming concepts have already made their way into modern cities. In recent years, the creation of urban farming technologies could be seen in many big cities around the world, including Singapore. To increase the resilience of the food supply chain, Singapore has developed its own approach by identifying them as "three food baskets": Diversifying food sources, Growing locally, and Growing overseas. Through “growing locally”, the policy makers have formulated a "30 by 30" plan to build up Singapore’s food industry's capacity and to have 30% of Singapore's nutritional needs produced locally by 2030. To support Singapore’s mission and address challenges such as food security and sustainability, A*STAR’s Institute of High Performance Computing (IHPC) developed an intelligent urban farm model - iSAM that integrates lighting simulation, light recipe optimisation, and farm performance simulation and evaluation, helping farmers to study their crops’ health, lighting recommendations, and farm design optimisation for shorter growth and higher harvest cycle.


The integrated smart farm model (iSAM) currently includes the following four modules:
  • Layout based hybrid lighting farm design (FarmMAP): Propose farm design with light recipes in a dynamic environment with inputs of sensor data to evaluate the potential yield of the farm 
  • Lighting modelling and simulation (GrowLITE): Optimise lighting environment for farms and assess and recommend hybrid lighting strategies for greenhouses by considering farm structure, light recipe of crops, energy usage, etc.  
  • Crop growth cycle simulation (CropSIM): Simulate the full crop growth cycle with light recipes of different plant stages and crop growth rates under dynamic environmental conditions
  • Biomass prediction and optimal light recipe (HarvestAI): Train artificial intelligence (AI) model to predict the biomass of plants and recommend the light recipe for optimal plant growth 

The Science Behind

iSAM is an intelligent platform that could help modern farmers study their crops’ health, lighting recommendations, and farm design optimisation for shorter growth and higher harvest cycle. 

Farm Design Modelling

Fig 1. Farm design modelling and optimisation

FarmMAP creates and visualises the farm object and layout. By serialising the farm objects, it communicates with other modules for evaluating the light distribution and intensity inside the farm and the potential crop yield, to achieve an optimal farm design (Fig 1).

light distribution and intensity on crops
Fig 2: Light distribution and intensity on crops of a typical day

GrowLITE simulates the light spectrums in terms of wavelength ranges and intensity by considering natural light and/or artificial light to understand the light distribution and the amount of light a crop in the farm can receive (Fig 2).

light conditions and crop light recipes estimation
Fig 3: Crop growth cycle simulation

CropSIM evaluates the farm design by estimating the monthly or yearly yield under various environmental factors, including light conditions and crop light recipes through an agent-based simulation module (Fig 3).

biomass prediction and optimisation of light recipe

Fig 4: AI models for biomass prediction and optimisation of light recipe

HarvestAI analyses the experimental and/or existing crop growth data and predicts crop growth rates. It also recommends the optimal light recipe using an AI-based recommendation engine (Fig 4). 

The developed technologies can equip the farmers to best utilise the limited natural resources such as land and space in the urban environment to increase crop production. In addition, with better yield and using lesser energy by applying optimal lighting recipe, it results in lower carbon footprints.

Industry Applications

The iSAM platform is a smart farm modelling platform for optimising the crop yield and potentially helping balance the energy utilisation and emission. 

  • Agritech – Indoor farms, greenhouse, rooftop farms
  • Farm Design – Energy efficiency, layout optimisation
  • Energy – Energy computation and optimisation, net zero energy usage
  • Recycling and sustainability - alternative energy applications