Indonesian Journal of Electrical Engineering and Computer Science Vol. 16, No. 3, December 2019, pp. 1311~1319 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v16.i3.pp1311-1319  1311 Journal homepage: http://iaescore.com/journals/index.php/ijeecs Algorithm for assessing forest stand productivity index using leaf area index Faid Abdul Manan, Muhammad Buce Saleh, I Nengah Surati Jaya, Uus Saepul Mukarom Department of Forest Management, Faculty of Forestry IPB University, Indonesia Article Info ABSTRACT Article history: Received Feb 11, 2019 Revised Jul 2, 2019 Accepted Jul 28, 2019 This paper describes a development of an algorithm for assessing stand productivity by considering the stand variables. Forest stand productivity is one of the crucial information that required to establish the business plan for unit management at the beginning of forest planning activity. The main study objective is to find out the most significant and accurate variable combination to be used for assessing the forest stand productivity, as well as to develop productivity estimation model based on leaf area index. The study found the best stand variable combination in assessing stand productivity were density of poles (X2), volume of commercial tree having diameter at breast height (dbh) 20-40 cm (X16), basal area of commercial tree of dbh >40 cm (X20) with Kappa Accuracy of 90.56% for classifying into 5 stand productivity classes. It was recognized that the examined algorithm provides excellent accuracy of 100% when the stand productivity was classified into only 3 classes. The best model for assessing the stand productivity index with leaf area index is y = 0.6214x - 0.9928 with R2= 0.71, where y is productivity index and x is leaf area index. Keywords: Discriminant analysis Estimator model Forest dimension Forest inventory Hemispherrical photography Copyright © 2019 Institute of Advanced Engineering and Science. All rights reserved. Corresponding Author: I NengahSurati Jaya, Department of Forest Management, Bogor Agricultural University, Jl. Raya Dramaga, Bogor 16680, West Java, Indonesia. Email: ins-jaya@apps.ipb.ac.id 1. INTRODUCTION Estimating forest stand productivity with traditional method is a challenging task, since it should be laborious, time consuming and costly. However, implementation of terrestrial forest inventory activities for determining the forest productivity in the field are often constrained by the availability of the human resources, cost and time. Thus, the terrestrial-based method is mostly difficult to be implemented in large area coverage. Many studies [1-13] proven that the variation of the standing stock is affected by its biological and physical factors that may be verified with leaf area index (LAI), density, basal area, volume, biomass, silviculture system, and site quality. The leaf area index is defined as one side of the leaf from the total area of leaf tissue (m 2 ) per unit of land surface area without considering the shape of the leaf [14, 15]. The leaf area index (LAI) is important for studying the crown structure and it is widely used to describe photosynthesis and transpiration of plant canopy surfaces [16]. LAI also emerged as a key factor in determining the impact of global climate change of forest ecosystem [17] and highly correlated with growth and forest products [18]. The LAI hold a significant role in determining the productivity of forest stands through the role of radiation interception [19]. In addition, LAI is one of the important variables in the functional and physiological models of plants [20], large scale remote sensing models [21], as well as large models ecosystem productivity models [22]. Now, there are several tools have been developed for estimating LAI and assess crown structure with image analysis through light transmittance measurements, one of which is digital hemispherical photography