diff --git a/README.md b/README.md
index c323832..a2becb7 100644
--- a/README.md
+++ b/README.md
@@ -65,13 +65,15 @@ algorithms in Rust and benchmark them. Let the battle begin.
 
 - `single thread`: No multi-threading or any optimization of any kind. Used as
   a reference.
+- `lpt`: Manual multi-threading. Tasks are sorted so the shortest ones are
+  executed first. This is the [longest-processing-time-first algorithm][lpt].
 - `rayon`: Multi-threading provided by the `rayon` crate, but no optimization
   of any kind. Used as a reference.
 - `rayon sorted`: Multi-threading provided by the `rayon` crate. Tasks are
   sorted so the shortest ones are executed first.
 - `rayon sorted reverse`: Multi-threading provided by the `rayon` crate. Tasks
   are sorted so the longest ones are executed first. This should behave as an
-  [LPT][lpt].
+  LPT.
 - `rayon sorted reverse`: Similar to `rayon sorted reverse`, but tasks are
   sorted from shortest to longest and are then iterated in the reverse order.
 
diff --git a/benches/job_scheduling.rs b/benches/job_scheduling.rs
index 8fb62c1..8d8ef53 100644
--- a/benches/job_scheduling.rs
+++ b/benches/job_scheduling.rs
@@ -6,6 +6,7 @@ use rust_job_scheduling::Task;
 const TEST_DATA: &[(&str, &[Task])] = &[("10 tasks", T_10), ("500 tasks", T_500)];
 const TEST_FUNCS: &[(&str, &dyn Fn(&[Task]))] = &[
 	("single thread", &single_thread),
+	("lpt", &lpt),
 	("rayon", &rayon),
 	("rayon sorted", &rayon_sorted),
 	("rayon sorted reverse", &rayon_sorted_reverse),
diff --git a/src/algorithm.rs b/src/algorithm.rs
index f75d047..e46e616 100644
--- a/src/algorithm.rs
+++ b/src/algorithm.rs
@@ -1,5 +1,7 @@
+mod lpt;
 mod rayon;
 mod single_thread;
 
+pub use lpt::lpt;
 pub use rayon::*;
 pub use single_thread::single_thread;
diff --git a/src/algorithm/lpt.rs b/src/algorithm/lpt.rs
new file mode 100644
index 0000000..237d493
--- /dev/null
+++ b/src/algorithm/lpt.rs
@@ -0,0 +1,33 @@
+use crate::Task;
+use std::sync::{Arc, Mutex};
+use std::thread;
+
+pub fn lpt(data: &[Task]) {
+	let mut data = data.to_vec();
+	data.sort();
+
+	let data_mt = Arc::new(Mutex::new(data));
+	let nb_threads = thread::available_parallelism()
+		.map(|n| n.get())
+		.unwrap_or(1);
+	let mut handles = Vec::with_capacity(nb_threads);
+
+	for _ in 0..nb_threads {
+		let dt = data_mt.clone();
+		let h = thread::spawn(move || loop {
+			let mut data = dt.lock().unwrap();
+			let task = match data.pop() {
+				Some(t) => t,
+				None => {
+					break;
+				}
+			};
+			std::mem::drop(data);
+			task.execute();
+		});
+		handles.push(h);
+	}
+	for h in handles {
+		let _ = h.join();
+	}
+}