From a440752c4e4f7bd40d906d5d8ae30f060ddbe4ba Mon Sep 17 00:00:00 2001
From: Alexander Hess <alexander@webartifex.biz>
Date: Fri, 9 Oct 2020 14:48:29 +0200
Subject: [PATCH] Rephrase abstract

- streamline first half
- add "on a grid"
- add note on predictive routing applications
- mention managerial findings and winning forecasting method
  in second half
---
 tex/meta.tex | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/tex/meta.tex b/tex/meta.tex
index d944b33..8296540 100644
--- a/tex/meta.tex
+++ b/tex/meta.tex
@@ -23,18 +23,18 @@ Emails:
 
 \begin{abstract}
 Meal delivery platforms like Uber Eats shape the landscape in cities around the world.
-This paper addresses forecasting demand into the short-term future.
+This paper addresses forecasting demand on a grid into the short-term future,
+    enabling, for example, predictive routing applications.
 We propose an approach incorporating
-    both classical forecasting
-    and machine learning methods.
-Model evaluation and selection is adapted to demand typical for such a platform
-    (i.e., intermittent with a double-seasonal pattern).
-The results of an empirical study with a European meal delivery service show
-    that machine learning models become competitive
-    once the average daily demand passes a threshold.
-As a main contribution, the paper explains
-    how a forecasting system must be set up
-    to enable predictive routing.
+    both classical forecasting and machine learning methods
+    and adapt model evaluation and selection to typical demand:
+        intermittent with a double-seasonal pattern.
+An empirical study shows that
+    an exponential smoothing based method trained on past demand data alone
+        achieves optimal accuracy,
+    if at least two months are on record.
+With only a shorter demand history,
+    machine learning methods and real-time data may improve prediction.
 \end{abstract}
 
 \begin{keyword}