diff --git a/stable/concourse/Chart.yaml b/stable/concourse/Chart.yaml
index 2f0467c55b..9bfe93c40e 100644
--- a/stable/concourse/Chart.yaml
+++ b/stable/concourse/Chart.yaml
@@ -1,5 +1,5 @@
 name: concourse
-version: 1.3.2
+version: 1.4.0
 appVersion: 3.10.0
 description: Concourse is a simple and scalable CI system.
 icon: https://avatars1.githubusercontent.com/u/7809479
diff --git a/stable/concourse/README.md b/stable/concourse/README.md
index 21287fddd2..545be71b67 100644
--- a/stable/concourse/README.md
+++ b/stable/concourse/README.md
@@ -129,6 +129,7 @@ The following table lists the configurable parameters of the Concourse chart and
 | `worker.fatalErrors` | Newline delimited strings which, when logged, should trigger a restart of the worker | *See [values.yaml](values.yaml)* |
 | `worker.updateStrategy` | `OnDelete` or `RollingUpdate` (requires Kubernetes >= 1.7) | `RollingUpdate` |
 | `worker.podManagementPolicy` | `OrderedReady` or `Parallel` (requires Kubernetes >= 1.7) | `Parallel` |
+| `worker.hardAntiAffinity` | Should the workers be forced (as opposed to preferred) to be on different nodes? | `false` |
 | `persistence.enabled` | Enable Concourse persistence using Persistent Volume Claims | `true` |
 | `persistence.worker.storageClass` | Concourse Worker Persistent Volume Storage Class | `generic` |
 | `persistence.worker.accessMode` | Concourse Worker Persistent Volume Access Mode | `ReadWriteOnce` |
diff --git a/stable/concourse/templates/worker-statefulset.yaml b/stable/concourse/templates/worker-statefulset.yaml
index b33a7b607d..693fd0bd75 100644
--- a/stable/concourse/templates/worker-statefulset.yaml
+++ b/stable/concourse/templates/worker-statefulset.yaml
@@ -116,6 +116,14 @@ spec:
 {{ toYaml .Values.worker.additionalAffinities | indent 8 }}
 {{- end }}
         podAntiAffinity:
+          {{- if .Values.worker.hardAntiAffinity }}
+          requiredDuringSchedulingIgnoredDuringExecution:
+          - labelSelector:
+              matchLabels:
+                app: {{ template "concourse.worker.fullname" . }}
+                release: {{ .Release.Name | quote }}
+            topologyKey: kubernetes.io/hostname
+          {{- else }}
           preferredDuringSchedulingIgnoredDuringExecution:
           - weight: 100
             podAffinityTerm:
@@ -124,6 +132,7 @@ spec:
                 matchLabels:
                   app: {{ template "concourse.worker.fullname" . }}
                   release: {{ .Release.Name | quote }}
+          {{- end }}
       volumes:
         - name: concourse-keys
           secret:
diff --git a/stable/concourse/values.yaml b/stable/concourse/values.yaml
index 4ff3474afe..390afc59c3 100644
--- a/stable/concourse/values.yaml
+++ b/stable/concourse/values.yaml
@@ -342,6 +342,11 @@ worker:
   #               values:
   #                 - "true"
 
+  ## Whether the workers should be forced to run on separate nodes.
+  ## This is accomplished by setting their AntiAffinity with requiredDuringSchedulingIgnoredDuringExecution as opposed to preferred
+  ## Ref: https://kubernetes.io/docs/concepts/configuration/assign-pod-node/#inter-pod-affinity-and-anti-affinity-beta-feature
+  hardAntiAffinity: false
+
   ## Tolerations for the worker nodes.
   ## Ref: https://kubernetes.io/docs/concepts/configuration/taint-and-toleration/
   tolerations: []