add better pic

Author: yoelcortes

docs/images/UML.png

@@ -11,8 +11,28 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "TEA/LCA is a powerful tool to understand the potential sustainability of a technology. But it can be extremely difficult to navigate uncertainties in design decisions (e.g., refinery size and location), market variability, and technological performance. Evaluating just one representative scenario (under a single set of assumptions) gives an incomplete picture that is not conclusive. This is especially true for conceptual and early-stage technologies, which have higher levels of uncertainty. We need to pair TEA/LCA with rigorous uncertainty/sensitivity analyses to explore the landscape of potential outcomes, identify representative scenarios (through sensitivity analysis), and establish technological performance targets. To learn more about expediting RD&D through modeling, we recommend reading on [quantitive sustainable design (QSD) methodology](https://pubs.rsc.org/en/content/articlelanding/2022/ew/d2ew00431c).\n",
+    "TEA/LCA is a powerful tool to understand the potential sustainability of a technology. But it can be extremely difficult to navigate uncertainties in design decisions (e.g., refinery size and location), market variability, and technological performance. Evaluating just one representative scenario (under a single set of assumptions) gives an incomplete picture that is not conclusive. This is especially true for conceptual and early-stage technologies, which have higher levels of uncertainty. We need to pair TEA/LCA with rigorous uncertainty/sensitivity analyses to explore the landscape of potential outcomes, identify representative scenarios (through sensitivity analysis), and establish technological performance targets. To learn more about expediting RD&D through modeling, we recommend reading on [quantitive sustainable design (QSD) methodology](https://pubs.rsc.org/en/content/articlelanding/2022/ew/d2ew00431c)."
+   ]
+  },
+  {
+   "cell_type": "raw",
+   "metadata": {},
+   "source": [
+    ".. figure:: model_UML_light.png\n",
+    "   :figwidth: 60%\n",
+    "   :class: only-light\n",
+    "   :align: center\n",
     "\n",
+    ".. figure:: model_UML_dark.png\n",
+    "   :figwidth: 60%\n",
+    "   :class: only-dark\n",
+    "   :align: center"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
     "As one of its central features, BioSTEAM streamlines TEA/LCA with rigorous uncertainty/sensitivity analyses to empower researchers with the ability to navigate uncertainty and guide RD&D. Using NREL's model for cellulosic ethanol production from cornstover as a case study, this tutorial will demonstrate how to construct a model and perform Monte Carlo-based uncertainty/sensitivity analyses to establish potential targets for improvement. "
    ]
   },